Search is not available for this dataset
article
stringlengths 4.36k
149k
| summary
stringlengths 32
3.35k
| section_headings
sequencelengths 1
91
| keywords
sequencelengths 0
141
| year
stringclasses 13
values | title
stringlengths 20
281
|
|---|---|---|---|---|---|
Previous studies suggest that factual learning , that is , learning from obtained outcomes , is biased , such that participants preferentially take into account positive , as compared to negative , prediction errors . However , whether or not the prediction error valence also affects counterfactual learning , that is , learning from forgone outcomes , is unknown . To address this question , we analysed the performance of two groups of participants on reinforcement learning tasks using a computational model that was adapted to test if prediction error valence influences learning . We carried out two experiments: in the factual learning experiment , participants learned from partial feedback ( i . e . , the outcome of the chosen option only ) ; in the counterfactual learning experiment , participants learned from complete feedback information ( i . e . , the outcomes of both the chosen and unchosen option were displayed ) . In the factual learning experiment , we replicated previous findings of a valence-induced bias , whereby participants learned preferentially from positive , relative to negative , prediction errors . In contrast , for counterfactual learning , we found the opposite valence-induced bias: negative prediction errors were preferentially taken into account , relative to positive ones . When considering valence-induced bias in the context of both factual and counterfactual learning , it appears that people tend to preferentially take into account information that confirms their current choice .
Goal-directed behaviour is composed of two core components [1]: one component is the decision-making process that starts with representing the available options and terminates in selecting the option with the highest expected value; the second component is reinforcement learning ( RL ) , through which outcomes are used to refine value expectations , in order to improve decision-making . Human decision-making is subject to biases ( i . e . deviations from the normative prescriptions ) , such as the framing effect [2] . While the investigation of decision-making biases has a long history in economics and psychology , learning biases have been much less systematically investigated [3] . This is surprising as most of the decisions we deal with in everyday life are experience-based and choice contexts are recurrent , thus allowing learning to occur and therefore influencing future decision-making . In addition , it is important to investigate learning biases as there is evidence that RL processes play a role in psychiatric conditions and maladaptive economic behaviour [4 , 5] . Standard RL algorithms learn action-outcome associations directly from obtained outcomes on a trial and error basis [6] . We refer to this direct form of learning as “factual learning” . Despite the fact that standard models , built around the notion of computational and statistical optimality , prescribe that an agent should learn equally well from positive and negative obtained outcomes [7–9] , previous studies have consistently shown that humans display a significant valence-induced bias . This bias generally goes in the direction of preferential learning from positive , compared to negative , outcome prediction errors [10–14] . This learning asymmetry could represent a RL counterpart of the “good news/bad news” effect that is observed for probabilistic reasoning [15] . However , human RL cannot be reduced simply to learning from obtained outcomes . Other sources of information can be successfully integrated in order to improve performance and RL has a multi-modular structure [16] . Amongst the more sophisticated learning processes that have already been demonstrated in humans is counterfactual learning . Counterfactual learning refers to the ability to learn from forgone outcomes ( i . e . the outcomes of the option ( s ) that were not chosen ) [17 , 18] . Whether or not a valence-induced bias also affects counterfactual learning remains unknown . To address this question , we ran two experiments involving instrumental learning and computational model-based analyses . Two groups of healthy adults performed variants of a repeated two-armed bandit task involving probabilistic outcomes [19 , 20] ( Fig 1A ) . We analysed the data using a modified Rescorla-Wagner model that assumes different learning rates for positive and negative , and factual and counterfactual , prediction errors ( Fig 1B ) [21 , 22] . The first experiment aimed to replicate previous findings of a “positivity bias” at the level of factual learning . In this first experiment , participants were presented only with the obtained outcome ( chosen outcome: RC; Fig 1A ) [10] . In the second experiment , in order to investigate whether or not counterfactual learning rates are also affected by the valence of prediction errors , we used a variant of the same instrumental learning task , in which participants were also presented with the forgone outcome ( unchosen outcome: RU; Fig 1B ) . Our design allowed us to test three competing hypotheses concerning the effect of valence on counterfactual learning ( Fig 2A ) . The first hypothesis–“no bias”—was that unlike factual learning , counterfactual learning would be unbiased . The second hypothesis , —“positivity bias”—was that factual and counterfactual learning would present the same valence-induced bias , such that positive counterfactual prediction errors would be more likely to be taken into account than negative counterfactual prediction errors . In this scenario , factual and counterfactual learning biases would be consequences of a more general positivity bias , in which positive prediction errors have a greater impact on learning , regardless of whether the option was chosen or not . Finally , the third hypothesis–“confirmation bias”—was that valence would affect factual and counterfactual learning in opposing directions , such that negative unchosen prediction errors would be more likely to be taken into account than positive unchosen prediction errors . In this scenario , factual and counterfactual learning biases would be consequences of a more general confirmation bias , in which outcomes that support the current choice are preferentially taken into account .
To investigate both factual and counterfactual reinforcement learning biases , we designed an instrumental task based on a previous paradigm , in which we showed a significant positivity bias in factual learning [10] . Here , we used two variants of the task , which differed in that the task used in Experiment 1 involved participants ( N = 20 ) being shown only the outcome of their chosen option , whereas in Experiment 2 ( N = 20 ) the outcome of the unchosen option was also displayed ( Fig 1A ) . To test our hypotheses concerning valence-induced learning biases ( Fig 2A ) we fitted the data with a Rescorla-Wagner model assuming different learning rates for positive and negative outcomes , which respectively elicit positive and negative prediction errors ( Fig 1B ) . The algorithm used to explain Experiment 1 data involved two learning rates for obtained outcomes ( αc+ and αc− for positive and negative prediction errors of the obtained outcomes , respectively ) . In addition to the obtained outcome learning rates , the algorithm used to explain Experiment 2 data also involved two learning rates for forgone outcomes ( αu+ and αu− for positive and negative prediction errors of the forgone outcomes , respectively ) . Replicating previous findings , in Experiment 1 we found that the positive factual learning rate ( αc+ ) was significantly higher than the negative one ( αc−; T ( 19 ) = 2 . 4; P = 0 . 03 ) ( Fig 2B , left ) . In Experiment 2 , we analysed learning rates using a repeated-measure ANOVA with prediction error valence ( positive or negative ) and prediction error type ( factual or counterfactual ) as within-subjects factors . Falsifying the “positivity bias” hypothesis , the ANOVA revealed no main effect of prediction error valence ( F ( 1 , 19 ) = 0 . 2; P>0 . 6 ) . We also did not find any effect of prediction error type , indicating that , on average , factual and counterfactual learning were similar ( F ( 1 , 19 ) = 0 . 5; P>0 . 4 ) . Consistent with the “confirmation bias” hypothesis , we found a significant interaction between valence and type ( F ( 1 , 19 ) = 119 . 2; P = 1 . 3e-9 ) . Post-hoc tests indicated that the interaction was driven by effects of valence on both factual ( αc+>αc−; T ( 19 ) = 3 . 6; P = 0 . 0017 ) and counterfactual learning rates ( αu−>αu+; T ( 19 ) = 6 . 2; P = 5 . 8e-06 ) ( Fig 2B , right ) . To verify the robustness of this result in the context of different reward contingencies , we analysed learning rates in each task condition separately . In both experiments , our task included three different conditions ( S1 Fig ) : a “Symmetric” condition , in which both options were associated with a 50% chance of getting a reward; an “Asymmetric” condition , in which one option was associated with a 75% chance of getting a reward , whereas the other option was associated with only a 25% chance; and a “Reversal” condition , in which one option was initially associated with a 83% chance of getting a reward and the other option was associated with a 17% chance of getting a reward , but after 12 trials the reward contingencies were reversed . For Experiment 1 , we analysed factual learning rates using a repeated-measure ANOVA with prediction error valence ( positive and negative ) and task condition ( Symmetric , Asymmetric and Reversal ) as within-subjects factors ( S1B Fig ) . Confirming the aggregate result , the ANOVA showed a significant main effect of valence ( F ( 1 , 19 ) = 26 . 4 , P = 5 . 8e-5 ) , but no effect of condition ( F ( 2 , 38 ) = 0 . 7 , P>0 . 5 ) , and , crucially , no valence by condition interaction ( F ( 2 , 38 ) = 0 . 8 , P>0 . 4 ) . For Experiment 2 , we analysed factual and counterfactual learning rates using a repeated-measure ANOVA with prediction error valence ( positive and negative ) , prediction error type ( factual or counterfactual ) and condition ( Symmetric , Asymmetric and Reversal ) as within-subjects factors ( S1C Fig ) . Confirming the aggregate result , the ANOVA showed no effect of prediction error type ( F ( 1 , 19 ) = 0 . 0 , P>0 . 9 ) , no effect of valence ( F ( 1 , 19 ) = 0 . 3 , P>0 . 5 ) , but a significant valence by type interaction ( F ( 1 , 19 ) = 162 . 9 , P = 9 . 1e-11 ) . We also found an effect of condition ( F ( 2 , 38 ) = 5 . 1 , P = 0 . 01 ) , reflecting lower average learning rates in the Reversal compared to the Asymmetric condition ( T ( 19 ) = 2 . 99; P = 0 . 007 ) , which was not modulated by valence ( F ( 2 , 38 ) = 0 . 2 , P>0 . 7 ) , or type ( F ( 2 , 38 ) = 1 . 2 , P>0 . 3 ) . The three-way interaction was not significant ( F ( 2 , 38 ) = 1 . 8 , P> . 1 ) , indicating that learning biases were robust across different task contingencies . To further test our hypotheses and verify theparsimony of our findings , we ran a model comparison analysis including the ‘Full’ model ( i . e . , the model with four learning rates; Fig 1C , right ) and reduced , alternative versions of it ( Fig 3A ) . The first alternative model was obtained by reducing the number of learning rates along the dimension of the outcome type ( factual or counterfactual ) . This ‘Information’ model has only two learning rates: one for the obtained outcomes ( αC ) and another for the forgone outcomes ( αU ) . The second alternative model was obtained by reducing the number of learning rates along the dimension of the outcome valence ( positive or negative ) . This ‘Valence’ model has only two learning rates ( one for the positive outcomes ( α+ ) and another for the negative outcomes ( α- ) ) and should win according to the “positivity bias” hypothesis . Finally , the third alternative model was obtained by reducing the learning rate as a function of the outcome event being confirmatory ( positive obtained or negative forgone ) or disconfirmatory ( negative obtained or positive forgone ) . This ‘Confirmation’ model has only two learning rates ( one for confirmatory outcomes ( αCON ) and another for the disconfirmatory outcomes ( αDIS ) ) and should win according to the “confirmation bias” hypothesis . Bayesian Information Criterion ( BIC ) analysis indicated that the ‘Full’ model better accounted for the data compared to both the ‘Information’ and the ‘Valence’ models ( both comparisons: T ( 19 ) >4 . 2; P<0 . 0005; Table 1 ) . However the ‘Confirmation’ model better accounted for the data compared to the ‘Full’ model ( T ( 19 ) = 9 . 9; P = 6 . 4e-9 ) . The posterior probability ( PP ) of belonging to each model , calculated for each subject , ( i . e . , the averaged individual model attributions ) of the ‘Confirmation’ model was higher than chance ( . 0 . 25 for a model space including 4 models; T ( 19 ) = 13 . 5; P = 3 . 3e-11 ) and higher than the posterior probability all the other models ( all comparisons: T ( 19 ) >9 . 0; P<2 . 1e-8 ) ( Fig 3B ) . The learning rate for confirmatory outcomes was significantly higher than that for disconfirmatory outcomes ( αCON>αDIS; T ( 19 ) = 11 . 7; P = 3 . 9e-10 ) ( Fig 3C ) . These results support the “confirmation bias” hypothesis and further indicate that , at least at the behavioural level , chosen and unchosen outcomes may be processed by the same learning systems . To evaluate the capacity of our models to reproduce the learning curves , we plotted and analysed the trial-by-trial model estimates of choice probabilities ( Fig 4 ) [23] . The model estimates were generated using the best fitting set of parameters for each individual and model . In the Symmetric condition ( where there is no correct response ) , we considered the preferred option choice rate ( i . e . , the option/symbol that was chosen more than >50% ) . In the Asymmetric condition we considered the correct choice rate . In the Reversal condition ( where the correct response is reversed after the first half of the trials ) we considered the choice rate of the initially more advantageous option ( i . e . , the correct option during the first half ) . Qualitative observation of the learning curves indicated that the biased models ( Experiment 1: αc+≠αc−; Experiment 2: αCON≠αDIS ) tended to reproduce the learning curves more closely . To quantify this , we compared the average square distance between the biased and the unbiased models ( Experiment 1: αc+=αc−; Experiment 2: αCON = αDIS ) . We found that the square distance was shorter for the biased models compared to the unbiased models in both experiments ( Experiment 1: 0 . 074 vs . 0 . 085 , T ( 19 ) = 3 . 5 P = 0 . 0022; Experiment 2: 0 . 056 vs . 0 . 064 , T ( 19 ) = 3 . 5 P = 0 . 0016 ) . We calculated the Pearson correlation between the parameters ( Fig 5A ) and found no significant correlation when correcting for multiple comparisons ( corrected P value = 0 . 05÷6 = 0 . 008; lowest uncorrected P value = 0 . 01 , highest P2 = 0 . 30 ) . The correlation between αCON and αDIS was weak , but positive , which rules out the possibility that the significant difference between these two learning rates was driven by an anti-correlation induced by the model fitting procedure . We then applied the same model fitting procedure to the synthetic datasets and calculated the correlation between the true and the retrieved parameters ( Fig 5B ) . We found that , on average , all parameters in both experiments were well recovered ( 0 . 70 ≤ R ≤ 0 . 89 ) and that our model fitting procedure introduced no spurious correlations between the other parameters ( |R| ≤ 0 . 5 ) . We also checked the parameter recovery for discrete sets of parameter values ( S2 & S3 Figs ) . For Experiment 1 , we simulated unbiased ( αc+=αc− ) and biased ( αc+>αc− ) participants . For Experiment 2 , we simulated unbiased ( αc+=αc− and αu+=αu− ) , semi-biased ( αc+>αc− and αu+=αu− ) and biased ( αc+>αc− and αu+>αu− ) participants . We simulated N = 100 virtual participants per set of parameters . The results of these analyses are presented in the supplementary materials and confirm the capacity of our parameter optimisation procedure to correctly recover the true parameters , regardless of the presence ( or absence ) of learning rate biases . To investigate the behavioural consequences of the learning biases , we median-split the participants from each experiment into two groups according to their normalised learning rate differences . We reasoned that the effects of learning biases on behavioural performance could be highlighted by comparing participants who differed in the extent they expressed the bias itself . Experiment 1 participants were split according to their normalised factual learning rate bias: ( αc+−αc− ) / ( αc++αc− ) , from which we obtained a high ( M = 0 . 76±0 . 05 ) and a low bias ( M = 0 . 11±0 . 14 ) group . Experiment 2 participants were split according their normalised confirmation bias: [ ( αc+−αc− ) − ( αu++αu− ) ]/ ( αc++αc−+αu++αu− ) , from which we also obtained a high bias group ( M = 0 . 72±0 . 04 ) and a low bias group ( M = 0 . 36±0 . 04 ) . From the Symmetric condition we extracted preferred choice rate as a dependent variable , which was the choice rate of the most frequently chosen option ( i . e . the option that was chosen on >50% of trials ) ( Fig 6A ) . We hypothesised that higher biases were associated with an increased tendency to develop a preferred choice , even in the absence of a “correct” option , which naturally emerges from overweighting positive factual ( and/or negative counterfactual ) outcomes , as observed in our previous study [10] . We submitted the preferred choice rate to an ANOVA with experiment ( 1 vs . 2 ) and bias level ( high vs . low ) as between-subjects factors . The ANOVA showed a significant main effect of bias level ( F ( 1 , 36 ) = 8 . 8 , P = 0 . 006 ) . There was no significant main effect of experiment ( F ( 1 , 36 ) = 0 . 6 , P>0 . 6 ) and no significant interaction between experiment and bias level ( F ( 1 , 36 ) = 0 . 3 , P>0 . 5 ) . Replicating previous findings , the main effect of bias level was driven by higher preferred choice rate in the high , compared to the low bias group in both Experiment 1 ( T ( 18 ) = 1 . 8 P = 0 . 08 ) and Experiment 2 ( T ( 18 ) = 2 . 3 P = 0 . 03 ) ( Fig 6B & 6C ) . From the remaining conditions we extracted the correct choice rate , which was the choice rate of the most frequently rewarded option . In the Reversal condition , correct choice rate was split across the first half of the trial ( i . e . , before the reversal of the contingencies ) and second half ( i . e . , after the reversal of the contingencies ) ( Fig 6A ) . We hypothesised that in the second half of the Reversal condition , where correct choice rate depends on un-learning previous associations based on negative factual prediction errors ( and positive counterfactual prediction errors , in Experiment 2 ) , high bias subjects will display reduced performance . We submitted the correct choice rate to a mixed ANOVA with experiment ( 1 vs . 2 ) and bias group ( high vs . low ) as between-subjects factors , and condition ( Asymmetric , Reversal: first half , and Reversal: second half ) as a within-subjects factor . There was a main effect of experiment ( F ( 1 , 36 ) = 4 . 1 , P = 0 . 05 ) , indicating that correct choice rate was higher in Experiment 2 than Experiment 1 , which is consistent with previous studies showing that counterfactual feedback enhances learning[20 , 24] . We also found a significant effect of bias level ( F ( 1 , 36 ) = 10 . 8 , P = 0 . 002 ) , a significant effect of condition ( F ( 2 , 72 ) = 99 . 5 , P = 2 . 0e-16 ) , and a significant bias level by condition interaction ( F ( 2 , 72 ) = 9 . 6 , P = 0 . 0002 ) . Indeed , in both experiments , the correct choice rate in the second half of the Reversal condition was lower in the high bias compared to the low bias group ( Experiment 1: T ( 18 ) = 3 . 9 P = 0 . 0003; Experiment 2: T ( 18 ) = 2 . 5 P = 0 . 02 ) ( Fig 6B & 6C ) . Importantly , we found that the temperature did not differ between low and high bias subjects in Experiment 1 ( low vs . high: 3 . 38±0 . 82 vs . 3 . 78±0 . 67; T ( 18 ) = 0 . 4 , P = 0 . 7078 ) or in Experiment 2 ( low vs . high . 3 . 29±0 . 56 vs . 2 . 13±0 . 36; T ( 18 ) = 1 . 7 , P = 0 . 0973 ) . Of note , the difference in temperature goes in two different directions in the two experiments , whereas the behavioural effects ( i . e . , increased preferred response rate in the Symmetric condition and decreased performance in the second half of the Reversal condition ) go in the same direction . Finally , we used Pearson’s correlations to verify that the relevant results remained significant when assessed as continuous variables . As predicted , the normalised learning biases were significantly and positively correlated with the preferred choice rate in the Symmetric condition in both experiments ( Experiment 1: R = 0 . 54 , P = 0 . 013; Experiment 2: R = 0 . 46 , P = 0 . 040 ) . Similarly , the normalised learning biases were significantly and negatively correlated with the correct choice rate in the second half of the Reversal condition ( Experiment 1: R = -0 . 66 , P = 0 . 0015; Experiment 2: R = -0 . 47 , P = 0 . 033 ) .
Two groups of healthy adult participants performed two variants of an instrumental learning task , involving factual ( Experiment 1 ) and counterfactual ( Experiments 1 & 2 ) reinforcement learning . We found that prediction error valence biased factual and counterfactual learning in opposite directions . Replicating previous findings , we found that , when learning from obtained outcomes ( factual learning ) , the learning rate for positive prediction errors was higher than the learning rate for negative prediction errors . In contrast , when learning from forgone outcomes ( counterfactual learning ) , the learning rate for positive prediction errors was lower than that of negative prediction errors . This result proved stable across different reward contingency conditions and was further supported by model comparison analyses , which indicated that the most parsimonious model was a model with different learning rates for confirmatory and disconfirmatory events , regardless of outcome type ( factual or counterfactual ) and valence ( positive or negative ) . Finally , behavioural analyses showed that participants with a higher valence-induced learning bias displayed poorer learning performance , specifically when it was necessary to adjust their behaviour in response to a reversal of reward contingencies . These learning biases were therefore significantly associated with reduced learning performance and can be considered maladaptive in the context or our experimental tasks . Our results demonstrated a factual learning bias , which replicates previous findings by showing that , in simple instrumental learning tasks , participants preferentially learn from positive compared to negative prediction errors [11–13] . However , in contrast to previous studies , in which this learning bias had no negative impact on behavioural performance ( i . e . , correct choice rate and therefore final payoff ) , here we demonstrated that this learning bias is still present in situations in which it has a negative impact on performance . In fact , whereas low and high bias participants performed equally well in conditions with stable reward contingencies , in conditions with unstable reward contingencies we found that high bias participants showed a relatively reduced correct choice rate . When reward contingencies were changed , learning to successfully reverse the response in the second half of the trials was mainly driven by negative factual ( and positive counterfactual ) prediction errors . Thus in this case , participants displaying higher biases exhibited a lower correct choice rate . In other words , these learning biases significantly undermined participants’ capacity to flexibly adapt their behaviour in changing , uncertain environments . In addition to reduced reversal learning , and in accordance with a previous study [10] , another behavioural feature that distinguished higher and lower bias participants was the preferred response rate in the Symmetric condition . In the Symmetric condition , both cues had the same reward probabilities ( 50% ) , such that there was no intrinsic “correct” response . This allowed us to calculate the preferred response rate for each participant ( defined as the choice rate of the option most frequently selected by a given participant , i . e . the option selected in > 50% of trials ) . The preferred response rate can therefore be taken as a measure of the tendency to overestimate the value of one cue compared to the other , in the absence of actual outcome-based evidence . In both experiments , higher bias participants showed higher preferred response rates , a behavioural pattern that is consistent with an increased tendency to discount negative factual ( and positive counterfactual ) prediction errors . This can result in one considering a previously rewarded chosen option as better than it really is and an increased preference for this choice . Thus , these results illustrate that the higher the learning bias for a given participant , the higher his/her behavioural perseveration ( the tendency to repeat a previous choice ) , despite the possible acquisition of new evidence in the form of negative feedback . Previous studies have been unable to distinguish whether this valence-induced factual learning bias is a “positivity bias” or a “confirmation bias” . In other words , do participants preferentially learn from positive prediction errors because they are positively valenced or because the outcome confirms the choice they have just made ? To address this question we designed Experiment 2 in which , by including counterfactual feedback , we were able to separate the influence of prediction error valence ( positive vs . negative ) from the influence of prediction error type ( chosen vs . unchosen outcome ) . Crucially , whereas the two competing hypotheses ( “positivity bias” vs . “confirmation bias” ) predicted the same result concerning factual learning rates , they predicted opposite effects of valence on counterfactual learning rates . The results from Experiment 2 support the confirmation bias hypothesis: participants preferentially took into account the outcomes that confirmed their current behavioural policy ( positive chosen and negative unchosen outcomes ) and discounted the outcomes that contradicted it ( negative chosen and positive unchosen outcomes ) . Our results therefore support the idea that confirmation biases are pervasive in human cognition [25] . It should be noted that , from an orthodox Bayesian perspective , a confirmation bias would involve reinforcing one's own initial beliefs or preferences . Previous studies have investigated how prior information—in the form of explicit task instructions or advice—influences the learning of reinforcement statistics and have provided evidence of a confirmation bias [26–28] . However , consistent with our study , their computational and neural results suggest that this instruction-induced confirmation bias operates at the level of outcome processing and not at the level of initial preferences or at the level of the decision-making process [29 , 30] . Here , we take a slightly different perspective by extending the notion of confirmation bias to the implicit reinforcement of one's own current choice , by preferentially learning from desirable outcomes , independently from explicit prior information . We performed a learning rate analysis separately for each task condition and the results proved robust and were not driven by any particular reward contingency condition . Our results contrast with previous studies that have found learning rates adapt as a function of task contingencies , showing increases when task contingencies were unstable [31 , 32] . Several differences between these tasks and ours may explain this discrepancy . First , in previous studies , the stable and unstable phases were clearly separated , whereas in our design , participants were simultaneously tested in the three reward contingency conditions . Second , we did not explicitly tell participants to monitor the stability of the reward contingency . Finally , since in our task the Reversal condition represented only one quarter of the trials , participants may not have explicitly realised that changing learning rates were adaptive in some cases . To date , two different views of counterfactual learning have been proposed . According to one view , factual and counterfactual learning are underpinned by different systems that could be computationally and anatomically mapped onto subcortical , model-free modules , and prefrontal , model-based modules [17 , 18 , 33] . In contrast , according to another view , factual and counterfactual outcomes are processed by the same learning system , involving the dopaminergic nuclei and their projections [34–36] . Our dimensionality reduction model comparison result sheds new light on this debate . If the first view was correct , and factual and counterfactual learning are based on different systems , different learning rates for positive and negative prediction errors would have better accounted for the data ( the ‘Information’ model ) . In contrast , our results showed that the winning model was one in which the learning process was assumed to be different across desirable and undesirable outcomes , but shared across obtained and forgone outcomes ( as in the “Confirmation” model ) , This supports the second view that factual and counterfactual learning are different facets of the same system . Overall , we found that correct choice rate was higher in Experiment 2 than in Experiment 1 , indicating that the presence of complete feedback information improved performance . Previous literature in psychology and economics suggest that this beneficial effect of counterfactual information is conditional on the payoff structure of the task . Specifically , studies have shown that the presence of rare positive outcomes could impair performance in the presence of complete feedback [37–40] . Further research is needed to assess whether or not the learning biases we identified extend to these payoff schemes and how they relate to the observed performance impairment . Another series of studies in psychology and economics have used paradigms that dissociate information sampling ( i . e . , choosing an option to discover its value without getting the outcome ) from actual choice ( i . e . , choosing an option in order to obtain the associated outcome ) [3] . Other paradigms have been used to investigate learning from outcomes derived from choices performed by either a computer or another player ( i . e . , observational learning ) [41 , 42] . Future research should assess whether or not information sampling and observational learning present similar valence-induced learning biases . Why do these learning biases exist ? One possibility is that these learning biases arise from neurobiological constraints , which limit human learning capacity . However , we believe this interpretation is unlikely because we see no clear reason why such limits would differentially affect learning from positive and negative prediction errors . In other words , we would predict that neurobiological constraints on learning rate would limit all learning rates in a similar way and therefore not produce valence-induced learning asymmetries . A second possibility is that these learning biases are not maladaptive . For instance , it has been shown that in certain reward conditions agents displaying valence-induced learning biases may outperform unbiased agents [9] . Thus , a possible explanation for these learning biases is that they have been positively selected because they can be adaptive in the context of the natural environment in which the learning system evolved [43] . A third , intermediate possibility is that these learning biases can be maladaptive in the context of learning performance , but due to their adaptive effects in other domains of cognition , overall they have a net adaptive value . For example , these biases may also manifest as “self-serving” , choice-supportive biases , which result in individuals tending to ascribe success to their own abilities and efforts , but relatively tending to neglect their own failures [44] . Accordingly , we could speculate that these learning biases may help promote self-esteem and confidence , both of which have been associated with overall favourable real life outcomes [45] . In summary , by investigating both factual and counterfactual learning , the current experiments demonstrate that , when presented with new evidence , people tend to discard information that suggests they have made a mistake . This selective neglect of useful information may have adaptive value , by increasing self-confidence and self-esteem . However , this low level reinforcement-learning bias may represent a computational building block for higher level cognitive biases such as belief perseverance , that is , the phenomenon that beliefs are remarkably resilient in the face of empirical challenges that logically contradict them [46 , 47] .
The study included two experiments . Each experiment involved N = 20 participants ( Experiment 1: 7 males , mean age 23 . 9 ± 0 . 7; Experiment 2: 4 males , mean age 22 . 8 ± 0 . 7 ) . The local ethics committee approved the study . All participants gave written informed consent before inclusion in the study , which was carried out in accordance with the declaration of Helsinki ( 1964 , revised 2013 ) . The inclusion criteria were being older than 18 years and reporting no history of neurological or psychiatric disorders . Participants performed a probabilistic instrumental learning task based on previous studies [19 , 20] ( Fig 1A ) . Briefly , the task involved choosing between two cues that were presented in fixed pairs and therefore represented fixed choice contexts . Cues were associated with stationary outcome probabilities in three out of four contexts . In the remaining context the outcome probability was non-stationary . The possible outcomes were either winning or losing a point . To allow learning , each context was presented in 24 trials . Each session comprised the four learning contexts and therefore included 96 trials . The whole experiment involved two sessions , each including the same number of contexts and conditions , but a different set of stimuli . Thus , the total experiment included 192 trials . The four learning contexts ( i . e . fixed pairs of cues ) were divided in three conditions ( S1 Fig ) . In the “Symmetric” condition each cue was associated with a . 50 probability of winning one point . In the “Asymmetric” condition one cue was associated with a . 75 probability of winning a point and the other cue was associated with a . 25 probability of winning a point . The Asymmetric condition was implemented in two choice contexts in each session . Finally , in the “Reversal” condition one cue was associated with a . 83 probability of winning a point and the other cue was associated with a . 17 probability of winning a point during the first 12 trials , and these contingencies were reversed thereafter . We chose a bigger probability difference in the Reversal compared to the Asymmetric condition in order to ensure that participants were able to reach a plateau within the first 12 trials . Participants were encouraged to accumulate as many points as possible and were informed that some cues would result in winning more often than others . Participants were given no explicit information regarding reward probabilities , which they had to learn through trial and error . At each trial , after a fixation cross , the choice context was presented . Participants made their choice by pressing left or right arrow keys with their right hand ( the choice time was self-paced ) . The two experiments differed in the fact that in the Experiment 1 participants were only informed about the outcome of their own choice ( chosen outcome ) , whereas in the Experiment 2 participants were informed about both the obtained and the forgone outcome ( i . e . counterfactual feedback ) . In Experiment 1 positive outcomes were presented at the top and negative outcomes at the bottom of the screen . The participant was required to press the key corresponding to the position of the outcome on the screen ( top/bottom ) in order to move to the subsequent trial . In Experiment 2 the obtained outcomes were presented in the same place as the chosen cues and the forgone outcomes in the same place as the unchosen cues . To move to the subsequent trial , participants had to match the position of the outcome with a key press ( right/left ) . Importantly for our computational analyses , outcome probabilities ( although on average anti-correlated in the Asymmetric and Reversal conditions ) were truly independent across cues , so that in the Symmetric condition , in a given trial , the obtained and forgone outcomes were the same in 50% of trials; in the Asymmetric condition this was the case in 37 . 5% of trials; finally , in the Reversal condition this was the case in 28 . 2% of trials . We extracted the correct response rate , that is , the rate of the trials in which the participants chose the most rewarding response , from the Asymmetric and the Reversal conditions . The correct response rate in the Reversal condition was calculated separately for the two phases: before ( “first half” ) and after ( “second half” ) the contingency reversal . In the Symmetric condition , we calculated the so-called “preferred” response rate . The preferred response was defined as the most frequently chosen option , i . e . that chosen by the participant on more than 50% of the trials . This quantity was therefore , by definition , greater than 0 . 5 . To investigate the behavioural consequences of learning biases on performance , we median-split the participants from each experiment into two groups according to their normalised learning rate difference ( Experiment 1: ( αc+−αc− ) / ( αc++αc− ) ; Experiment 2: [ ( αc+−αc− ) − ( αu++αu− ) ]/ ( αc++αc−+αu++αu− ) ) , from which we obtained ‘low’ and ‘high’ bias participants[48] . The preferred response rate in the Symmetric condition was submitted to an ANOVA with experiment ( 1 vs . 2 ) and bias level ( high vs . low ) as between-subjects factors . The correct choice rate in the remaining conditions was submitted to an ANOVA with experiment ( 1 vs . 2 ) and bias level ( high vs . low ) as between-subjects factors and condition ( Asymmetric , Reversal first half and Reversal second half ) as within-subject factors . The effects of interest identified by the ANOVAs were also confirmed using Pearson’s correlations . We fitted the data with a standard Q-learning model , including different learning rates following positive and negative prediction errors and containing two different modules ( Fig 1C ) : a factual learning module to learn from chosen outcomes ( Rc ) and a counterfactual learning module to learn from unchosen outcomes ( Ru ) ( note that counterfactual learning applies only to Experiment 2 ) . For each pair of cues ( choice context ) , the model estimates the expected values of the two options ( Q-values ) . These Q-values essentially represent the expected reward obtained by choosing a particular option in a given context . In both experiments , Q-values were set at 0 before learning , corresponding to the a priori expectation of a 50% chance of winning 1 point , plus a 50% chance of losing 1 point . After every trial t , the value of the chosen option is updated according to the following rule ( factual learning module ) : Qc ( t+1 ) =Qc ( t ) +αc+ . PEc ( t ) ifPEc ( t ) >0αc− . PEc ( t ) ifPEc ( t ) <0 ( 1 ) In this first equation , PEc ( t ) is the prediction error of the chosen option , calculated as: PEc ( t ) =Rc ( t ) −Qc ( t ) , ( 2 ) where Rc ( t ) is the reward obtained as an outcome of choosing c at trial t . In other words , the prediction error PEc ( t ) is the difference between the expected outcome Qc ( t ) and the actual outcome Rc ( t ) . In Experiment 2 the unchosen option value is also updated according to the following rule ( counterfactual learning module ) : Qu ( t+1 ) =Qu ( t ) +αu+ . PEu ( t ) ifPEu ( t ) >0αu− . PEu ( t ) ifPEu ( t ) <0 ( 3 ) In this second equation , PEu ( t ) is the prediction error of the unchosen option , calculated as: PEu ( t ) =Ru ( t ) −Qu ( t ) , ( 4 ) where Ru ( t ) is the reward that could have been obtained as an outcome of having chosen u at trial t . In other words , the prediction error PEu ( t ) is the difference between the expected outcome Qu ( t ) and the actual outcome Ru ( t ) of the unchosen option . The learning rates αc+ and αu+ are scaling parameters that adjust the amplitude of value changes from one trial to the next when prediction errors of chosen and unchosen options , respectively , are positive ( when the actual reward R ( t ) is better than the expected reward Q ( t ) ) . The learning rates αc− and αu− do the same when prediction errors are negative . Thus , our model allows for the amplitude of value updates to be different following positive and negative prediction errors , and for both chosen and unchosen options . It therefore allows for the existence of valence-dependent learning biases . Finally , the probability ( or likelihood ) of selecting the chosen option was estimated with a soft-max rule as follows: Pc ( t ) =e ( Qc ( t ) *β ) / ( e ( Qc ( t ) *β ) +e ( Qu ( t ) *β ) ) . ( 5 ) This is a standard stochastic decision rule that calculates the probability of selecting one of a set of options according to their associated values . The temperature , β , is another scaling parameter that adjusts the stochasticity of decision-making . In addition to this ‘Full’ model , we also considered alternative versions with a reduced number of learning rates ( Fig 3A ) : the ‘Information’ model , where αc+=αc− and αu+=αu−; the ‘Valence’ model , where αc+=αu+ and αc−=αu−; and the ‘Confirmation’ model , where αc+=αu− and αc−=αu+ . For the model comparison , we also considered a very simple model ( the ‘One’ ) model , with only one learning rate ( αc+=αc−=αu+=αu− ) , and a ‘Perseveration’ model where an additional parameter ( –Inf < π < +Inf ) biases the decision-making process by increasing ( positive values ) or decreasing ( negative values ) the likelihood of repeating the same choice , regardless of the previous outcome ( Table 1 ) . In a first analysis , we optimised model parameters by minimising the negative log-likelihood of the data , given different parameter settings , using Matlab’s fmincon function ( ranges: 0<β<Infinite , and 0< αn<1 ) : LL=log ( P ( Data|Model ) ) ( 6 ) Negative log-likelihoods ( LL ) were used to compute the Bayesian information criterion ( BIC ) at the individual level ( random effects ) for each model , as follows: BIC=log ( ntrials ) *df+2*LL ( 7 ) BIC were compared between biased and unbiased models to verify that the utilisation of the biased model was justified , even accounting for its extra-complexity . As an approximation of the model evidence , individual BICs were fed into the mbb-vb-toolbox [49] , a procedure that estimates the exceedance probability and the model attributions for each model within a set of models , given the data gathered from all participants . Exceedance probability ( denoted XP ) is the probability that a given model fits the data better than all other models in the set , i . e . has the highest XP ( Table 1 ) . The toolbox also allows the estimation of the individual model attributions , i . e . the posterior probability ( PP ) , for each subject , of belonging to each model . The individual model attributions can be compared to chance ( defined as 1/the total number of models ) , compared to each other , and can also be averaged to obtain the model frequency for the population . In a second analysis , we optimised model parameters by minimising the logarithm of the Laplace approximation to the model evidence ( or log posterior probability: LPP ) : LPP=log ( P ( Data|Model , Parameters ) ) ( 8 ) Because LPP maximisation includes priors over the parameters ( temperature: gamma ( 1 . 2 , 5 ) ; learning rates beta ( 1 . 1 , 1 . 1 ) ) [50] , it avoids degenerate parameter estimates . Therefore , learning rate analyses have been performed on the values retrieved with this procedure . To avoid bias in learning rate comparisons , the same priors were used for all learning rates . In the main analysis , a single set of parameters was used to fit all conditions . In a control analysis , different sets of parameters were used to fit each condition ( “Symmetric” , “Asymmetric” and “Reversal” ) . To validate our results , and more specifically to verify that valence-induced differences in learning rates reflected true differences in learning , as opposed to an artefact of the parameter optimisation procedure , we checked the correlations between the free parameters ( Experiment 1: β , αc+ , αc−; Experiment 2: β , αCON , αDIS ) and the capacity of recovering the correct parameters using simulated datasets . To check the capacity of recovering the correct parameters using simulated datasets , we simulated performance on our behavioural task using virtual participants with parameters values corresponding to those retrieved from our experimental participants [23] . We simulated N = 100 virtual experiments . | While the investigation of decision-making biases has a long history in economics and psychology , learning biases have been much less systematically investigated . This is surprising as most of the choices we deal with in everyday life are recurrent , thus allowing learning to occur and therefore influencing future decision-making . Combining behavioural testing and computational modeling , here we show that the valence of an outcome biases both factual and counterfactual learning . When considering factual and counterfactual learning together , it appears that people tend to preferentially take into account information that confirms their current choice . Increasing our understanding of learning biases will enable the refinement of existing models of value-based decision-making . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] | [
"learning",
"decision",
"making",
"social",
"sciences",
"neuroscience",
"learning",
"and",
"memory",
"optimization",
"analysis",
"of",
"variance",
"cognitive",
"psychology",
"mathematics",
"statistics",
"(mathematics)",
"cognition",
"research",
"and",
"analysis",
"methods",
"learning",
"curves",
"human",
"learning",
"behavior",
"mathematical",
"and",
"statistical",
"techniques",
"economics",
"economic",
"history",
"psychology",
"biology",
"and",
"life",
"sciences",
"physical",
"sciences",
"cognitive",
"science",
"statistical",
"methods"
] | 2017 | Confirmation bias in human reinforcement learning: Evidence from counterfactual feedback processing |
Spinocerebellar Ataxia Type 2 ( SCA2 ) is caused by expansion of a polyglutamine encoding triplet repeat in the human ATXN2 gene beyond ( CAG ) 31 . This is thought to mediate toxic gain-of-function by protein aggregation and to affect RNA processing , resulting in degenerative processes affecting preferentially cerebellar neurons . As a faithful animal model , we generated a knock-in mouse replacing the single CAG of murine Atxn2 with CAG42 , a frequent patient genotype . This expansion size was inherited stably . The mice showed phenotypes with reduced weight and later motor incoordination . Although brain Atxn2 mRNA became elevated , soluble ATXN2 protein levels diminished over time , which might explain partial loss-of-function effects . Deficits in soluble ATXN2 protein correlated with the appearance of insoluble ATXN2 , a progressive feature in cerebellum possibly reflecting toxic gains-of-function . Since in vitro ATXN2 overexpression was known to reduce levels of its protein interactor PABPC1 , we studied expansion effects on PABPC1 . In cortex , PABPC1 transcript and soluble and insoluble protein levels were increased . In the more vulnerable cerebellum , the progressive insolubility of PABPC1 was accompanied by decreased soluble protein levels , with PABPC1 mRNA showing no compensatory increase . The sequestration of PABPC1 into insolubility by ATXN2 function gains was validated in human cell culture . To understand consequences on mRNA processing , transcriptome profiles at medium and old age in three different tissues were studied and demonstrated a selective induction of Fbxw8 in the old cerebellum . Fbxw8 is encoded next to the Atxn2 locus and was shown in vitro to decrease the level of expanded insoluble ATXN2 protein . In conclusion , our data support the concept that expanded ATXN2 undergoes progressive insolubility and affects PABPC1 by a toxic gain-of-function mechanism with tissue-specific effects , which may be partially alleviated by the induction of FBXW8 .
Spinocerebellar Ataxia Type 2 ( SCA2 ) is one of 9 currently known inherited neurodegenerative diseases ( e . g . Huntington's disease , SCA1 , SCA3 , SBMA ) that are caused by an expanded CAG trinucleotide repeat within the coding region of the disease gene , which expands its size from generation to generation and is translated to a polyglutamine ( polyQ ) domain [1]–[4] . More than 90% of the human population carry a repeat size of 22–23 triplets in the Ataxin-2 ( ATXN2 ) gene [3] , while alleles between 27 and 33 are considered intermediate size expansions and were recently shown to result in a higher risk for related neurodegenerative diseases such as ALS and Parkinsonism [5]–[7] . Individuals having a polyQ-repeat of 32 CAGs or more may develop SCA2 [8] , with larger repeat sizes resulting in a more severe disease course and earlier manifestation . Clinical disease onset occurs usually in late adult life [9]–[11] . The clinical manifestations of all Spinocerebellar Ataxias are similar , making the molecular genotyping indispensable to establish diagnosis [8] . Increased appetite with subsequent loss of subcutaneous fat and of weight are peripheral tissue features in other polyQ neurodegenerative diseases such as Huntington's disease [12] , [13] and are also common in SCA2 patients [8] , [14] . The primary characteristic manifestation in SCA2 patients reflects cerebellar incoordination signs like gait/stance/limb ataxia , dysarthria , dysmetria , adiadochokinesia , action tremor and hypotonia . Besides , the thalamus , brainstem , cranial nerves , spinal cord and muscles are affected early on , leading to a relatively characteristic manifestation with reduced saccade velocity , altered sleep , reduced deep tendon reflexes and cramps [15]–[23] . Early involvement of the midbrain can also lead to a manifestation as Parkinsonism [24] , [25] . Degeneration of the basal ganglia and the precerebellar nuclei is also present , while the affection of the cerebral cortex occurs relatively late [15] , [26] . Parkinsonism is also frequent among the few patients with SCA2 homozygosity , where generally the same vulnerability pattern was observed , but relatively early affection of the retina , the oculomotor , reticulotegmental , facial , lateral vestibular , raphe interpositus nuclei in the brainstem and some pyramidal cells in the primary motor cortex was reported [27]–[29] . Therapy is mostly palliative , but a surprising effect of deep brain stimulation on tremor was reported in one case [30] . In the end stage of disease , patients die mostly from respiratory failure [8] . Microscopic SCA2 findings in the cerebellum include early loss of the large Purkinje neurons , preceded by a reduction in their dendritic arborisation and in the thickness of the molecular layer where the synaptic contacts and neuronal processes are located [15] . While pathological intranuclear inclusion bodies ( NIIs ) containing polyQ aggregates in SCA2 are detectable only in a minority of brainstem neurons and never in cerebellar Purkinje neurons , cytoplasmic aggregates of ATXN2 were reported in SCA2 patient brains and in a mouse mutant with Purkinje-cell specific maximal transgenic overexpression of ATXN2 [31]–[33] , raising the question to what extent the formation of inclusion bodies containing the insoluble disease protein is the driving force in the neurodegenerative process of SCA2 . The expanded polyglutamine domain in other disease proteins is thought to mediate pathogenesis mainly by a toxic gain-of-function mechanism , in particular through aggregation of the mutant protein [34] . Partial loss-of-function effects have also been noted for polyQ disease proteins where the physiological function can already be quantified [35]–[37] . The few facts known about the function of ATXN2 suggest that it exerts its neurodegenerative effects as part of protein-RNA complexes [5] . ATXN2 is involved to some extent in trophic signalling [38] . Sequence analysis of ATXN2 revealed a PAM2 motif that mediates direct binding with the poly ( A ) -binding protein PABPC1 [39] , [40] and two Lsm motifs that may be involved in RNA processing . In the mammalian cell line HEK293 , the reduced ATXN2 levels lead to increased endogenous PABPC1 levels and vice versa [41] . ATXN2 protein associates with polyribosomes [40] and is mainly localized at the rough endoplasmic reticulum [42] . Furthermore , ATXN2 was found to be part of and necessary for the formation of stress granules ( SGs ) , distinct foci within the cytoplasm , where untranslated mRNAs are translationally inhibited during conditions of cell stress [39] , [41] . The expression of ATXN2 in the cerebellum is prominent in Purkinje neurons that comprise a small minority of all cerebellar cells , but are conspicuous due to their large cytoplasm and dendritic trees , and their exceptional content of ribosomal machinery easily visualized in Nissl stains . In cortex the expression of ATXN2 is widespread through neuron populations of several layers ( compare Allen brain atlas and reference [43] ) . As a faithful animal model of SCA2 , we now generated the first ATXN2 polyQ knock-in mouse . While mice usually have 1 CAG at the site of the human ATXN2 triplet repeat , the knock-in mice were designed to carry 42 CAGs in this locus . Knock-in mice have three distinct advantages over available models [44] . Firstly , the preferential vulnerability of specific tissues can be investigated in contrast to models employing heterologous promoters . For this aim we chose to compare the cerebellum as an early , severely affected tissue versus cerebral cortex where the neurons are lost only at late stages and where cognitive signs originate only in few patients . Secondly , in knock-in mice the contribution of partial loss-of-function versus toxic gain-of-function effects can be studied , in contrast to transgenic overexpressing animals or transfected cells that model SCA2 through elevated dosage of ATXN2 DNA , mRNA and protein . For this purpose , we performed a careful evaluation of mutants from different litters in comparison to their wild-type littermates , assessing systematically the endogenous transcript levels , soluble protein and insoluble protein levels as driven by the endogenous promoter . Thirdly , knock-in mice have the advantage over knock-in cells that disease progression and mutation effects can be studied over years . Towards this goal , we decided to perform longitudinal observation from early adult life ( age 6 weeks ) to multimorbid senescence ( age 21 months ) and to document the progressive effects of the ATXN2 expansion on PABPC1 . We chose PABPC1 as a direct protein interactor of ATXN2 , which relocalizes together with ATXN2 under cell stress from the rER to SGs , is controlled in its protein level by the physiological function of ATXN2 , and has a well studied cellular role in RNA processing . The data observed indicate selective and progressive pathology of cerebellar tissue regarding ATXN2 insolubility , PABPC1 deficiency , FBXW8 upregulation and locomotor impairment .
We generated a mouse line that expresses expanded Atxn2 under the control of the endogenous murine Atxn2 promoter ( Figures S1 and S2 , Table S1 ) . To confirm the successful homologous recombination , a PCR with CAG-repeat flanking primers was performed on DNA tail biopsies ( Figure 1A ) , the product sequence verified and the introduction of the CAG42-repeat confirmed . To test how the repeat is transmitted in the mice over successive generations , the PCR-products from mice across nine generations from WT , heterozygous ( CAG1/CAG42 ) and homozygous ( CAG42 ) animals were subjected to fragment analysis and their size was determined ( Figure 1B ) . All products had the exact same length . These data indicate that the knock-in mice have a repeat of 42 CAGs that is stably transmitted . To investigate changes in body weight which frequently accompany neurodegenerative diseases and in particular SCA2 , littermate mice with homozygous mutant or wild-type genotypes ( Figure S3 ) were weighed at regular intervals from postnatal day 10 onwards . Already at the age of 10 days , CAG42 mice had a significant body weight reduction of 19 . 27% compared to wild-type littermates ( p = 0 . 0002 ) ; a reduction of 13–22% remained significant throughout their life span ( 20 days: p = 0 . 0005; 6 weeks: p = 0 . 0015; 3 months: p = 0 . 0025; 6 months: p = 0 . 0196; 12 months: p = 0 . 0141; 18 months: p = 0 . 0265; 21 months: p = 0 . 0496 ) . Still , the weight gain was similar in CAG42 and WT . The body weight of CAG1/CAG42 was consistently , but non-significantly less than WT ( Figure 2A ) . The knock-in mice did not show overt ataxic behaviour during cage life . To assess the development of a cerebellar phenotype in more detail , animals were placed on an accelerating rotarod apparatus and their latency to fall was recorded . At 6 weeks , the first time point tested , the latency to fall was 1 . 2-fold increased ( p = 0 . 0096 ) in CAG42 mice in comparison to their wild-type littermates . CAG1/CAG42 mice were performing equally well as wild-types . At 3 , 6 and 12 months , both CAG42 and CAG1/CAG42 showed no difference to WT in their performance . At the age of 18 months the CAG42 mice showed a significantly shorter latency to fall ( 0 . 68-fold , p = 0 . 0296 ) from the rotarod , while at 21 months significance was just missed ( 0 . 65-fold reduction , p = 0 . 062 ) , probably due to the low number of animals remaining alive and available for analyses . Thus , CAG42 at early adult age performed better than WT , but at old age considerably worse in this test paradigm with particular sensitivity for cerebellar dysfunction , reflecting possibly an initial overcompensation followed by late-onset pathology . CAG1/CAG42 mice even at old age still performed similarly well as the WT mice ( Figure 2B ) , in keeping with other reports that mice homozygous for polyQ expansion disease show a stronger phenotype than heterozygous animals . Grip strength remained unchanged , indicating that motor neuron pathology and paralysis are absent ( data not shown ) . In addition , footprint analysis did not show any differences between WT and CAG1/CAG42 or CAG42 mice up to 21 months ( data not shown ) . As ATXN2 triplet repeat expansions in human can also lead to Parkinson syndrome with impaired spontaneous movement , the motor activity of mice in an open field was recorded . Neither horizontal/vertical activity , total distance , movement time , number of movements , number of stereotypy counts , margin/centre distance nor margin/centre time were consistently altered ( data not shown ) . Consequently , there was no motor hyperactivity to explain the reduced body weight of knock-in animals . Since the Atxn2 knock-out mice show excessive weight , the present data might be interpreted as evidence for a gain-of-function effect of the Atxn2-CAG42-knock-in in peripheral tissues . In summary , a permanent reduction of body weight and a late-onset impairment of rota rod performance were the only apparent phenotypes in CAG42 mice . In order to verify that the homologous recombination event does not interfere with Atxn2 expression and to assess the stability of the expanded transcript , quantitative real-time RT-PCR ( qPCR ) was performed in cerebellum and cerebral cortex . While the cerebellar Atxn2 mRNA levels at 6 weeks of age were unchanged , at 6 months ( p = 0 . 0146 ) and 18 months ( p = 0 . 0339 ) they were significantly elevated to 1 . 07-fold levels . In the cortex , the upregulation of Atxn2 mRNA expression was already significant at 6 weeks ( 1 . 29-fold , p = 0 . 0101 ) and stayed significant at 6 months ( 1 . 16-fold , p = 0 . 0005 ) and 18 months ( 1 . 16-fold , p = 0 . 0474 ) . The transcript levels of the Ataxin-2 interactor Pabpc1 remained unchanged at all time points in cerebellum . In contrast , in cortex an upregulation of Pabpc1 mRNA to 1 . 07-fold ( p = 0 . 0061 ) and 1 . 16-fold ( p = 0 . 0028 ) became apparent at 6 and 18 months , respectively ( Figure 3 ) . Thus , a mild but statistically significant elevation of Atxn2 transcript became apparent during adult life in cerebellum and even earlier in cortex , indicating that the selection marker remaining in the Atxn2 genomic locus does not impair Atxn2 transcription and that the presence of a CAG42 repeat in the Atxn2 transcript does not lead to its instability . Abnormally elevated Pabpc1 mRNA levels became apparent only in the cortex by 6 months , a noteworthy finding since previous in vitro transient transfection studies [41] found the ATXN2 loss-of function to increase PABPC1 levels . We investigated the stability and steady-state levels of expanded ATXN2 in extracts of proteins soluble in RIPA extraction buffer . Expanded ATXN2 with 42 glutamines displayed slowed electrophoretic mobility in comparison to wild-type ATXN2 ( Figure 4 ) . Both in cerebellum and cortex , a reduction of soluble ATXN2 was detectable in the CAG42 mice . The levels were diminished in the cerebellum consistently at 6 weeks ( 0 . 75-fold , p = 0 . 1008 ) , 6 months ( 0 . 77-fold , p = 0 . 0808 ) and significantly reduced at 18 months ( 0 . 62-fold , p = 0 . 0037 ) . The reduction of levels in the cortex missed significance at 6 weeks ( 0 . 86-fold , p = 0 . 1553 ) , but became significant both at 6 months ( 0 . 55-fold , p = 0 . 0046 ) and 18 months ( 0 . 75-fold , p = 0 . 0344 ) . As expected by the elevated Pabpc1 transcript levels , cortical soluble PABPC1 levels were elevated by 18 months ( 1 . 28-fold , p = 0 . 0293 ) . Curiously , however , cerebellar soluble PABPC1 was significantly reduced already at 6 weeks ( 0 . 82-fold , p = 0 . 034 ) and again at 18 months ( 0 . 75-fold , p = 0 . 0235 ) ( Figure 4 ) . Thus , a reduction of soluble Q42-ATXN2 protein in spite of elevated transcript levels was a consistent finding in both tissues at all ages , in keeping with a partial loss-of-function and explaining the elevated PABPC1 transcription in the cortex . However , selectively in the cerebellum PABPC1 soluble protein was also reduced in absence of any compensatory mRNA induction . To test whether increased insolubility of Q42-ATXN2 explains the effects observed , proteins from the pellet after extraction of the soluble proteins were further extracted with a harsher buffer from 6 , 12 and 24 months old mice . Detection with the anti-ATXN2 antibody failed , possibly due to epitope masking . Detection with the 1C2 antibody which recognizes polyQ tracts of at least 38 glutamines [45] was successful in revealing the expected band of correct size which was not apparent in wild-type tissue and exclusively appeared in mutant tissue . In the cerebellum a clear increase of insoluble Q42-ATXN2 was detected , which was significant at 12 ( 3 . 14-fold , p = 0 . 0096 ) and 24 months ( 4 . 28-fold , p = 0 . 0002 ) in comparison to 6 months Q42-ATXN2 . In the cortex , however , an increase was not observed . Also , insoluble PABPC1 protein levels in the cerebellum increased over time and were significantly elevated at 24 months both in comparison to 24 months old WT ( 1 . 72-fold , p = 0 . 0078 ) and to 6 months old CAG42 tissue ( 1 . 9-fold , p = 0 . 0357 ) . More insoluble PABPC1 was observed in the cortex , a bias which manifested as a trend in 6 months old mice ( 1 . 464-fold , p = 0 . 0812 ) and became significant by 24 months in comparison to the WT of the same age ( 1 . 98-fold , p = 0 . 02 ) and also in comparison to 6 months old CAG42 mice ( 2 . 036-fold , p<0 . 0001 ) . In wild-type animals the levels of insoluble PABPC1 levels remained similarly low over time ( Figure 5A ) . Thus , increased insolubility of Q42-ATXN2 , particularly in the old cerebellum , appears to sequester PABPC1 into insolubility and may explain the decreased levels of soluble ATXN2 and PABPC1 . In order to test the interaction between ATXN2 and PABPC1 in cerebellum with an independent biochemical method , ATXN2 was immunoprecipitated from wild-type , CAG42 and ATXN2-knock-out tissue and the co-immunoprecipitation of PABPC1 documented in immunoblots . Q42-ATXN2 pulled down slightly more of PABPC1 than wild-type ATXN2 ( Figure 5B ) . Beyond previous reports of association between the two tagged recombinant transfected proteins , our data demonstrate for the first time this interaction between the two endogenous proteins in mammalian tissue . Thus , Q42-ATXN2 is likely to sequester PABPC1 into insolubility through direct interaction . To investigate whether this insolubility process results in the formation of visible cytoplasmic aggregates containing ATXN2 , as described in affected neurons during late stages of SCA2 patients , immunohistochemistry on cerebellar brain sections was performed . At all ages tested , 7 , 14 and 24 months , 1C2-immunoreactivity was purely nuclear in wild-type mice , while the presence of Q42-ATXN2 resulted in additional cytoplasmic signals of Purkinje neurons , starting discretely by 14 months age and becoming stronger by 24 months ( Figure 6A–6F ) . The specific and selective ATXN2 immunoreactivity of Purkinje neurons could be successfully visualized by an unmasking technique , while remaining undetectable in knock-out tissue ( Figure S4 ) . This ATXN2 immunoreactivity of Purkinje cells revealed a diffuse cytoplasmic distribution in WT cerebellum , while it became discretely granular at 14 months in individual Purkinje neurons and markedly granular at 24 months in most Purkinje neurons of CAG42 mice ( Figure 6G–6L ) . In addition , the PABPC1-immunoreactivity was diffusely cytoplasmic in WT mice , but became discretely granular in some Purkinje neurons by 14 months and markedly granular in most Purkinje neurons by 24 months of age ( Figure 6M–6R ) . Nuclear inclusion bodies were never detected . A decrease of the molecular layer thickness or of the Purkinje cell number ( data not shown ) , or a reduction in the intensity of Calbindin immunoreactivity ( Figure S5 ) was not observed even at the age of 24 months , suggesting that a neurodegenerative process is not yet detectable at this age . Thus , the histological analysis supports the concept of progressive insolubility and aggregation of Q42-ATXN2 as well as PABPC1 in the cerebellar Purkinje neurons which are the prominent site of pathology in human SCA2 . To confirm in vitro that insoluble expanded ATXN2 influences the PABPC1 solubility , HeLa cells were transfected transiently with CAG22-ATXN2 and CAG74-ATXN2 constructs . Successful overexpression was verified by qPCR and Western blot . Over 200-fold elevated CAG22-ATXN2 and CAG74-ATXN2 mRNA levels were accompanied by reduced PABPC1 transcript levels ( approximately 0 . 5-fold , p = 0 . 0087 and p = 0 . 0041 ) ( Figure 7A ) , while at the protein level ( Figure 7B ) a 7 . 2-fold ( p = 0 . 0374 ) and 6 . 7-fold ( p = 0 . 1667 ) excess of soluble Q22-ATXN2 or Q74-ATXN2 was found , accompanied by a reduction of endogenous soluble PABPC1 levels to 0 . 65-fold ( p = 0 . 0138 and p = 0 . 002 ) . To assess the insoluble fractions of both proteins , the pellet after the extraction of soluble proteins was dissolved and re-extracted with more detergents . Overexpressed Q22-ATXN2 and Q74-ATXN2 were strongly detectable , while endogenous ATXN2 in the empty vector control transfection was practically undetectable . Endogenous PABPC1 levels in the insoluble protein fraction were increased to 3 . 2-fold after both Q22-ATXN2 ( p = 0 . 0048 ) and Q74-ATXN2 ( p = 0 . 0184 ) overexpression . These data corroborate our previous tissue data that insoluble ATXN2 of either normal or expanded size sequesters its interactor protein PABPC1 into insolubility and may thus contribute to the reduction of soluble PABPC1 levels . They are also in keeping with the previous in vitro reports [41] that the increase in soluble ATXN2 levels leads to reduced PABPC1 expression levels . Attempting a survey of the alterations in mRNA processing underlying this progressive pathology , we used cerebellum , brainstem and liver tissue from presymptomatic 6 months and symptomatic 18 months old WT and CAG42 mice ( 3 versus 3 animals per young group , 4 versus 4 per old group ) to perform Affymetrix microarray transcriptome studies at a genome-wide level . At the age of 6 months , statistical analysis with correction for multiple testing after Benjamini-Hochberg demonstrated that no mRNAs showed dysregulated levels with significance in cerebellum and brainstem , while only one gene was differentially regulated in liver . This gene Cyp4a14 , is represented only by one oligonucleotide spot on the microarray . The Cytochrome P-450 enzyme family has a known role in cholesterol biosynthesis and the Cyp4a14 induction is a known response of liver tissue to altered fat content [46] . Since fatty liver is a feature already documented and molecularly investigated in ATXN2-KO mice [47] , we did not investigate this finding further . At the age of 18 months , altered regulation was significant after Benjamini-Hochberg correction for 20 genes in cerebellum , 14 genes in brainstem and 30 genes in liver . Among those , the dysregulation of 5 genes was recognized consistently by more than one oligonucleotide microarray spot ( Table 1 ) . Upon independent validation of the findings by qPCR with commercial Taqman assays , the dysregulations of Acat1 , Ifi27l1 and Lgals1 could not be reproduced in tissue from different animals; in contrast , a significant downregulation of Adam1a in cerebellum to 0 . 71-fold ( p = 0 . 0007 ) and brainstem to 0 . 73-fold ( p = 0 . 001 ) as well as a significant upregulation of Fbxw8 to 2 . 38-fold ( p<0 . 0001 ) in cerebellum were confirmed in this approach . Adam1a encodes a disintegrin/metalloprotease and has an established role for spermatogenesis and fertilization [48] . Thus , its dysregulation might relate to the reduced fertility observed in ATXN2-KO mice [47] . Fbxw8 encodes a WD-40 domain containing member of the F-box protein family , which are substrate-recognition mediators within a SCF ( SKP1-CUL1-F-box ) -type ubiquitin-E3-ligase complex involved in the tagging of phosphorylated target proteins to destine them for degradation [49] . To assess whether Fbxw8 induction on transcript level results in elevated FBXW8 protein levels , Western blots in 18 months old cerebellum of WT and CAG42 mice were performed ( Figure 8A ) . A significant upregulation of FBXW8 to 1 . 75-fold ( p = 0 . 043 ) in CAG42 mice was documented . Thus , elevated Fbxw8 mRNA and protein levels characterize the cerebellar tissue of old CAG42 mice . To further examine the questions whether the selective upregulation of Fbxw8 in old CAG42 cerebellum reflects a compensatory cellular response to pathology and whether this is an efficient regulation to degrade expanded ATXN2 also in humans , we assessed the effects of elevated FBXW8 protein on the soluble and insoluble levels of human ATXN2 in vitro . HeLa cells were transiently single and double transfected with CAG22-ATXN2 , CAG74-ATXN2 and FBXW8 ( Figure 8B ) . Transfection only with CAG22-ATXN2 or CAG74-ATXN2 in the soluble phase led to a 72-fold increase of Q22-ATXN2 ( p = 0 . 0098 ) or a 67-fold increase of Q74-ATXN2 ( p = 0 . 056 ) protein compared to the control vector pCMV . Additional expression of the control vector pReceiver did not reduce Q22-ATXN2 or Q74-levels significantly ( 57-fold , p = 0 . 64 , and 40-fold p = 0 . 51 ) . A further overexpression of FBXW8 reduced Q74-ATXN2-levels by a factor 3 . 3 ( p = 0 . 23 ) , while it did not markedly change Q22-ATXN2 levels . In the insoluble fraction , transfection led to elevated Q22-ATXN2 ( 126-fold , p<0 . 0001 ) and Q74-ATXN2 ( 59-fold , p<0 . 0001 ) levels . Additional expression of the empty pReceiver vector reduced Q22-ATXN2 levels ( 82-fold , p = 0 . 036 ) and did not alter Q74-ATXN2 levels ( 57-fold , p = 0 . 93 ) . After double transfection with FBXW8 and Q22-ATXN2 or Q74-ATXN2 , the levels of ATXN2 were again significantly elevated ( 64-fold , p = 0 . 0002 and 18-fold , p = 0 . 005 , respectively ) compared to the FBXW8+pCMV control . This elevated level was significantly less high for expanded ATXN2 in comparison to the control double transfection with pReceiver and CAG74-ATXN2 ( p = 0 . 028 ) . Control immunodetection of the FBXW8 levels demonstrated the successful overexpression to result in detectable upregulation in soluble and insoluble fractions , and a curious increase of FBXW8 insoluble protein levels in the presence of ATXN2 . These in vitro human data indicate that the upregulation of FBXW8 reduces the insoluble levels selectively of expanded ATXN2 . Analogous double transfection experiments showed that also the insoluble fraction of Q41-ATXN2 becomes slightly reduced by FBXW8 , but this effect did not become significant in three independent experiments ( Figure S6 ) .
We have generated and characterized the first knock-in mouse modelling Spinocerebellar Ataxia Type 2 . Since SCA2 is a late-onset progressive disease with selective vulnerability of the cerebellar Purkinje neurons , it is interesting that similar temporal dynamics were observed for the insolubility of ATXN2 and for the accompanying sequestration of PABPC1 in the knock-in cerebellum . The progressively granular immunoreactivity of ATXN2 and PABPC1 in Purkinje neuron immunohistochemistry supports the concept of aggregate accumulation , and the in vitro and Co-IP data confirm that excess ATXN2 becomes insoluble and recruits its interactor PABPC1 into insolubility . Progressively granular immunoreactivity of both proteins becomes apparent in increasing numbers of cerebellar Purkinje neurons even before the onset of the selectively cerebellar motor deficits . Probably this pathological process of aggregation corresponds to the toxic gain-of-function which was shown to drive disease progression in other polyQ disorders . A gain-of-function of ATXN2 might also underlie the reduced weight of the knock-in animals , since ATXN2 knock-out mice are known to have excessive weight [47] . Specifically in cerebellar tissue , the insolubility and sequestration led to a deficiency in soluble PABPC1 levels , an effect that might result in impaired RNA processing as well as protein synthesis [50] and therefore possibly critical for the Purkinje neurons with their large ribosomal machinery . Indeed , these molecular abnormalities correlate temporally with the late-onset appearance of motor deficit . The depletion of PABPC1 was shown to prevent global protein synthesis and to promote cell death in mammalian cells [50] . PABPC1 binds to the 3′ poly ( A ) tail of mRNAs to mediate their circularization which precedes ribosome recruitment and translation initiation [51] . Furthermore , PABPC1 is important for the poly ( A ) shortening and nonsense-mediated decay of mRNAs in the cytoplasm [52] , [53] . During the formation of stress granules , PABPC1 and ATXN2 relocalize there together with translationally repressed mRNA complexes [41] . Furthermore , ATXN2 could play a role in mRNA splicing , since it contains Lsm domains and since it interacts with A2BP1/Fox1 which is a well established splicing modulator [54] , [55] . Finally , ATXN2 might modulate transcription itself , since it was shown to relocalize to the nucleus and control its own expression [56] . Thus , there are multiple ways how ATXN2 could modulate RNA processing , and indeed the confirmed role of ATXN2 expansions in interaction with other RNA binding proteins for motoneuron degeneration diseases [5] , [6] , [57] indicates this to be a crucial aspect of ATXN2 effects . Our present data support the concept that reduced levels of PABPC1 are associated with the cerebellar vulnerability in SCA2 , and that the pathogenesis might have similarities to the altered mRNA processing in ALS , FXTAS and SMA [58] . Unfortunately , the rates of mRNA translation and deadenylation cannot be directly quantified in tissue , so further mechanistic analyses will have to rely on in vitro studies . It is interesting to note that in the cerebellum the progressive PABPC1 insolubility and the ATXN2 gain-of-function led to decreased soluble PABPC1 levels that did not elicit a compensatory upregulation of PABPC1 transcript levels . This may simply be due to a maximal PABPC1 expression level in cerebellar cells that cannot be enhanced further . In contrast , in cortex tissue a similarly progressive PABPC1 sequestration appeared overcompensated in the face of elevated levels of PABPC1 transcript and soluble protein . We speculate that the cause of this upregulated cortical PABPC1 expression lies in a response to the deficiency in soluble ATXN2 , since PABPC1 levels were previously found in vitro to correlate indirectly with ATXN2 levels [41] . This partial loss of soluble ATXN2 was observed in spite of elevated mRNA levels in a quite constant manner in all ages and brain regions . This is particularly interesting , since in other polyQ neurodegenerative diseases some minor symptoms have clearly been attributed to a partial loss-of-function mechanism . For example , polyQ expansions in the androgen receptor lead to a partial loss-of-function of this protein with subsequent testicular feminisation of male patients , apart from the atrophy of spinobulbar motoneurons which contain polyQ aggregates [59] . Our investigation into consequences of ATXN2 expansions for mRNA levels in a genome-wide transcriptome survey demonstrated a surprising scarcity of effects , with the notable exception of the highly significant , reproducible and specific dysregulation in the expression of two ATXN2 neighbour genes , Adam1a and Fbxw8 . The murine Atxn2 gene lies on chromosome 5 F ( position 122 , 162–122 , 265 kbp ) , with Adam1a ( position 121 , 969–121 , 972 kbp ) being positioned at a distance of ∼190 kbp from Atxn2 , and Fbxw8 ( position 118 , 515–118 , 606 kbp ) being located at a distance of ∼3556 kbp from Atxn2 . Three different hypotheses may explain these transcription effects with specificity to the Atxn2 locus . Firstly , it is conceivable that the knock-in targeting approach has modified the Atxn2 locus , but the facts that the selection markers were almost completely removed and that this locus effect was not detectable in the transcriptome data at 6 months age argue against this notion . Secondly , notions have been discussed that the abnormal DNA structure of a ( CAG ) -expansion could exert a progressive influence on chromatin structure [60]; indeed , polyglutamine expansions of the SCA7 disease protein , which is a component of a histone acetyltransferase complex , was shown to result in aberrant chromatin [61] . However , such effects were not confined to the surrounding locus and represented random toxic side-effects on unrelated proteins rather than an influence on the disease protein and its function . Thirdly , it is possible that the genes that surround the Atxn2 locus are interacting in a common pathway and that feedback-mechanisms regulate their expression in dependence on ATXN2 function . Our in vitro data that Fbxw8 upregulation may selectively reduce the levels of insoluble , expanded ATXN2 argues in favour of the latter hypothesis . A compensatory cellular effort that degrades the expanded and toxic ATXN2 and counteracts its gain-of-function would help to minimize the alterations of interactors with downstream consequences , thus postponing the onset of pathology; therefore FBXW8 represents a promising molecular target for neuroprotective approaches . FBXW8 was reported to control growth in parallel to modulating insulin-like growth factor binding protein 1 levels [62] . As FBXW8 null mice are growth retarded [62] , the induction of FBXW8 at old age in the knock-in tissue could again be interpreted as an effort to compensate the weight reduction caused by CAG42-ATXN2 . Furthermore , FBXW8 modifies neuronal dendrite formation [63] . It is also noteworthy that ATXN2 was observed to interact with the neuroprotective ubiquitin E3 ligase PARKIN [64] , and that FBXW7 ( SEL-10 ) is also known to interact with PARKIN [49] . It goes beyond the scope of this manuscript to test whether FBXW8 or other members of this protein family may also contribute to the degradation of other polyglutamine-expansion disease proteins like huntingtin and ataxin-1 . Overall , the early induction of Cyp14a4 and the later induction of Fbxw8 are probably part of the cellular stress response . In order to study further progression of the neurodegenerative process beyond such early compensatory efforts , and to understand later SCA2 and ALS stages , ATXN2 knock-in mice with larger CAG expansions or with CAA interruptions may have to be studied . In conclusion , our first report of a knock-in mouse with CAG42 to model SCA2 confirms that progressive insolubility of ATXN2 particularly in the cerebellum is the molecular feature which correlates best with the late-onset motor incoordination . The subsequent sequestration of the interactor protein PABPC1 into insolubility provides a first insight from the analysis of brain tissue for the recent focus how SCA2 and ALS pathogenesis might affect RNA processing [65] , [66] . The cerebellar tissue differed from cortical tissue by the lack of a compensatory upregulation of PABPC1 expression . Specifically in cerebellar tissue at old age , an upregulation of Fbxw8 levels was observed , which may contribute to counteract the ATXN2 gain-of-function according to our human in vitro studies . These findings may lead us to understand the tissue specificity in this disease process . We believe that this knock-in mouse will constitute a valuable tool for pathogenesis research in early stages of SCA2 and possibly also ALS , given the scarce availability of human autopsy brain tissues . Although the ATXN2 triplet repeat is transmitted stably across generations in mouse in contrast to humans , this may be an advantage for the long term comparability of data from this knock-in model .
Atxn2-CAG42-knock-in mice were generated by modifying the previously described pKO-Sca2-vector [47] . Exon 1 was deleted via the unique restriction sites SgrAI and Eco47III and replaced by a synthesized exon 1 fragment containing the CAG42-repeat ( custom-made by Geneart , Regensburg ) at the amino acid position glutamine 156 , resulting in the targeting vector NOW1-HR . The sequence verified vector was electroporated into 129Sv/Pas ES cells to allow for homologous recombination at the endogenous Ataxin-2 locus . Its integration was verified using PCR . The correct 3′ integration was verified using a forward primer hybridizing in the neomycin selection cassette and a reverse primer downstream of the targeting vector homology sequence . Verification of the correct 5′ integration was achieved with a primer pair flanking the loxP site . The forward primer was located upstream of the long homology arm of the targeting vector and the reverse primer hybridized upstream of the distal loxP site within the 5′ homology arm of the targeting vector . Additionally , the integration of the CAG repeat insertion was verified using primers flanking the CAG repeat . Further validation was performed by Southern blot analysis ( Figure S1 ) . Removal of the FRT flanked neomycin resistance cassette was performed via Flp-mediated excision and its success again verified using PCR and Southern blot analysis ( data not shown ) . Correctly targeted ES cell clones were injected into C57BL/6 blastocysts . This work was outsourced to Genoway ( Lyon , France ) . For primer sequences see Tables S2 , S3 , S4 , S5 . DNA from tail biopsies was isolated and the genotyping PCR was performed . TaKaRa LA Taq-Polymerase ( Takara Bio Inc . , Japan ) was used to amplify the neomycin cassette excised locus with the primer pair NOW1-K2 5′-TGAGTTGACTCCACAGGGAGGTGAGC-3′ and NOW1-H2 5′-CCATCTCGCCAGCCCGTAAGATTC-3′ flanking this site . The conditions were: initial denaturation at 94°C-3′ , followed by 30 cycles of 94°C-15″ denaturation , 68°C-4′ annealing and elongation and a final elongation step at 68°C-9′ . The wild-type ( WT ) allele is predicted to yield an amplification product of 793 bp and the knock-in ( CAG42 ) allele one of 984 bp , while heterozygous ( CAG1/CAG42 ) mice show products of both sizes ( Figure S3 ) . Mice were housed in accordance with the German Animal Welfare Act , the Council Directive of 24 November 1986 ( 86/609/EWG ) with Annex II and the ETS123 ( European Convention for the Protection of Vertebrate Animals ) at the FELASA-certified Central Animal Facility ( ZFE ) of the Frankfurt University Medical School . Mice were backcrossed from a mixed 129Sv/Pas×C57BL/6 for at least 8 generations into the C57BL/6 strain . Littermates derived from heterozygous matings were used for all experiments . For behaviour observations , male and female mice were used in similar proportions . Mice were weighed before behavioural testing . The motor performance of mice was assessed using an accelerating rotarod apparatus ( Ugo Basile , Comerio , Robert & Jones , model 7650 ) . Mice from three founder lines were placed at different ages on the accelerating rod ( from 4–40 rpm ) and latency to fall was recorded during the 5 minute trial . Grip strength was assessed by measuring the peak force of the fore limbs in 10 trials per mouse on an electronic grip strength meter ( TSE , Bad Homburg ) . Footprints were evaluated by painting the hind limbs of mice with a non-toxic ink . The mice were allowed to walk through a paper lined tunnel ( height: 6 cm×width: 9 cm×length: 40 cm ) and step length , gait width , alternation coefficient and linear movement were analyzed on the basis of the footprints as described previously [67] . The spontaneous motor activity of the mice was recorded by a Versamax animal activity monitor ( Accuscan , Columbus ) . The mouse was placed into the 20×20 cm arena and its activity recorded during the 5 minute trial . The determination of the CAG repeat length was done using fragment analysis . The CAG repeat was amplified with TaKaRa LA Taq-Polymerase ( Takara Bio Inc . , Japan ) from DNA from tail biopsies using a 5′-FAM-labeled forward primer 5′-CCCCGCCCGGCGTGCGAGCCGGTGTAT-3′ and the reverse primer 5′-CGGGCTTGCGGCCAGTGG-3′ under the following conditions: 96°C-4′ , followed by 38 cycles of 94°C-1′ , 60°C-1′ and 72°C-1′ with a final elongation step at 72°C-7′ . The determination of the fragment length was outsourced to GENterprise GENOMICS ( Mainz , Germany ) . Peak Scanner Software 1 . 0 ( Applied Biosystems ) was used to determine the exact PCR product size using the GS500 size standard . After cervical dislocation the brain was removed , cerebellum and cortex were dissected and quickly frozen in liquid nitrogen . RNA extraction from tissue was performed with Trizol reagent ( Invitrogen ) and from cells using the RNeasy Mini Kit ( Qiagen ) according to manufacturers' instructions . For expression studies , 1 µg of total RNA was digested with DNaseI Amplification Grade ( Invitrogen ) and reversely transcribed using SuperScript III Reverse Transcriptase ( Invitrogen ) . Expression levels were investigated with a StepOnePlus Real-Time PCR System ( Applied Biosystems ) . cDNA from 25 ng RNA were used in each PCR reaction with the following TaqMan Assays ( Applied Biosystems ) : Atxn2 ( Mm01199894_m1 , Hs00268077_m1 ) , Pabpc1 ( Mm00849569_s1 , Hs00743792_s1 ) , Acat1 ( Mm00507463_m1 ) , Adam1a ( Mm02581738_s1 ) , Fbxw8 ( Mm00554876_m1 ) , Ifi27l1 ( Mm00835449_g1 ) , Lgals1 ( Mm00839408_g1 ) and Tbp ( Mm00446973_m1 , Hs99999910_m1 ) as endogenous control . The PCR conditions were 50°C-2′ , 95°C-10′ followed by 40 cycles of 95°C-15″ and 60°C-60″ . The gene expression data were analyzed according to the 2−ΔΔCt method [68] . 50 mg of tissue were homogenized with a motor pestle in 10 vol . RIPA buffer [50 mM Tris-HCl ( pH 8 . 0 ) , 150 mM NaCl , 1 mM EDTA , 1 mM EGTA , 1% Igepal CA-630 ( Sigma ) , 0 . 5% sodium deoxycholate , 0 . 1% SDS , 1 mM PMSF and one tablet Complete Protease Inhibitor Cocktail ( Roche ) ] followed by 15 minutes incubation on ice . After centrifugation at 16 , 000×g at 4°C for 20 minutes , the supernatant ( soluble fraction ) was preserved and either 1/2 vol . Urea/SDS-buffer [8 M Urea , 5% SDS , 1 mM PMSF , Complete protease inhibitor cocktail ( Roche ) ] for tissue extraction or 1/2 vol . 2× SDS-lysis buffer [137 mM Tris-HCl , pH 6 . 8; 4% SDS; 20% glycerol] for cell extraction was added to the pellet , sonicated and centrifuged at full-speed for 10 minutes . The supernatant represented the insoluble fraction . Extracted proteins were processed directly without freezing them . Protein concentration was determined using the BCA protein assay kit ( Thermo Scientific ) . 10 or 20 µg of proteins were loaded and separated on 8% polyacrylamide gels and transferred to PVDF membranes . The membrane was blocked in 5% milk powder , incubated with primary antibodies against Ataxin-2 ( 1∶500 , BD Transduction Laboratories ) , PABPC1 ( 1∶1000 , Abcam ) , 1C2 ( 1∶1000 , Millipore ) , FBXW8 ( 1∶750 , Sigma ) and β-Actin ( 1∶10 , 000 , Sigma ) and visualized using ECL method ( Pierce ) . Densitometric analysis was carried out using the ImageJ software . The cerebellum of perfused ( 4% PFA ) mouse brains was cut on a microtome into 5 µm thick sagittal sections . For antibody staining , sections were immersed for 30 minutes in 10% methanol , 3% H2O2 and 50 mM Tris ( pH 7 . 6 ) and subsequently washed for three times in Tris . Sections were blocked for 90 minutes with 0 . 25% Triton X , 0 . 1 M D-Lysine and 10% Tris-BSA . Blocking before Ataxin-2 staining was modified using goat-serum in 0 . 1 M PBS . Incubation with the primary antibodies against Ataxin-2 ( BD Transduction laboratories , 1∶50 ) , PABPC1 ( Cell Signaling , 1∶50 ) , 1C2 ( 1∶1000 , Millipore ) or Calbindin ( 1∶3000 ) lasted 18 h at room temperature . The biotinylated secondary antibody ( 1∶200 ) was applied for 90 minutes followed by incubation for 2 h in the avidin-biotin-peroxidase complex ( 1∶100 in Tris , ABC-Elite , Vector Laboratories ) . Finally , sections were incubated in Tris with 0 . 07% DAB and 0 . 001% H2O2 . For Ataxin-2 staining sections were pretreated for 30 sec at 125°C in bull's eye decloaker solution in a decloaker chamber ( Decloaker , Biocare Medical ) ; for PABPC1 staining , sections were pretreated with citrate buffer; for 1C2 staining the sections were incubated in the microwave 3 times for 10 minutes in Tris ( pH 9 . 0 ) , followed by a 3 minute incubation step in 98% formic acid; for Calbindin staining the sections were pretreated with trypsin . Determination of the molecular layer thickness , Purkinje cell number and intensity of Calbindin immunoreactivity were done by ImageJ in 2 mice/genotype with at least 3 sections/animal and at least 10 microscopic fields/section . Approximately 50 mg of cerebellar tissue from 6 months old animals were lysed in 10× vol . lysis buffer ( 10 mM HEPES , 10 mM KCl , 5 mM MgCl2 , 0 . 1% Igepal CA-630 ( Sigma ) , protease inhibitors ) , incubated for 10 minutes at 4°C and subsequently centrifuged for 20 minutes at 16 , 000 g . 200 µg of soluble proteins derived from the supernatant were incubated overnight under rotation with anti-ATXN2 antibody . Beads ( Protein A/G-agarose , Santa Cruz ) were pre-treated overnight with washing buffer ( 0 . 2 NaCl , 1% gelatine , 0 . 05% NaN3 , 50 mM Tris , 0 . 1% Triton ) to avoid unspecific protein binding . Antibody-protein complexes were precipitated with beads under rotation for one hour at 4°C . Beads were sedimented by centrifugation and washed for 4 times in PBS including protease inhibitors and then subjected to Western blotting . HeLa cells were cultured in MEM medium ( Invitrogen ) with 10% FCS , 1% NEAA and 1% HEPES and seeded the day before the experiment . 1 µg of the plasmids pCMV-Myc , ATXN2 ( CAG22 ) -Myc , ATXN2 ( CAG74 ) -Myc [38] , FBXW8-pReceiver-M55 ( with Cherry-tag ) and pReceiver-M56 ( with Cherry-tag ) ( Imagenes ) were used for the transfection with Effectene Transfection Reagent ( Qiagen ) according to manufacturer's protocol . Cerebellum , brainstem and liver tissue from 6 and 18 months old WT and CAG42 mice were sent to the MFT Services ( Tübingen , Germany ) for analysis . The RNA was isolated , its quality documented ( one young CAG42 liver sample had to be eliminated from analysis ) and 100 ng of total RNA was amplified with the Affymetrix 3′ IVT Express Kit , labeled and hybridized onto the mouse specific MOE430 2 . 0 Gene Chip ( Affymetrix , Santa Clara , CA , USA ) that detects 39 , 000 transcripts and variants corresponding to 34 , 000 mouse genes . Scanning of the arrays was performed in the Gene Chip Scanner 3000 and the raw data obtained with the AGCC 3 . 0 software . Further analysis was done using the Bioconductor package ( www . bioconductor . org ) . RMA normalization was applied and calculation of differentially expressed transcripts was done by compiling a linear model . F-statistics was applied ( empirical Bayes model ) and the p-values obtained were further corrected for multiple testing with the “Benjamini-Hochberg” test . P-values<0 . 05 were considered significant . Genes that were detected within their exons and that were dysregulated consistently according to more than one spot were prioritized for further analysis . The original data from 18 months old animals have been deposited in NCBI's Gene Expression Omnibus [69] and are accessible through GEO Series accession number GSE39640 ( http://www . ncbi . nlm . nih . gov/geo/query/acc . cgi ? acc=GSE39640 ) ( NCBI tracking system #16608301 ) . The GraphPad Prism software version 4 . 03 ( 2005 ) and MS Excel 2007 ( Microsoft ) were used to perform unpaired Student's t-test . Error bars indicate SEM . Values p<0 . 05 were considered significant and marked with asterisks p<0 . 05 * , p<0 . 01 ** , p<0 . 001 *** . | Frequent age-associated neurodegenerative disorders like Alzheimer's , Parkinson's , and Lou Gehrig's disease are being elucidated molecularly by studying rare heritable variants . Various hereditary neurodegenerative disorders are caused by polyglutamine expansions in different proteins . In spite of this common pathogenesis and the pathological aggregation of most affected proteins , investigators were puzzled that the pattern of affected neuron population varies and that molecular mechanisms seem different between such disorders . The polyglutamine expansions in the Ataxin-2 ( ATXN2 ) protein are exceptional in view of the lack of aggregate clumps in nuclei of affected Purkinje neurons and well documented alterations of RNA processing in the resulting disorders SCA2 and ALS . Here , as a faithful disease model and to overcome the unavailability of autopsied patient brain tissues , we generated and characterized an ATXN2-CAG42-knock-in mouse mutant . Our data show that the unspecific , chronically present mutation leads to progressive insolubility and to reduced soluble levels of the disease protein and of an interactor protein , which modulates RNA processing . Compensatory efforts are particularly weak in vulnerable tissue . They appear to include the increased degradation of the toxic disease protein by FBXW8 . Thus the link between protein and RNA pathology becomes clear , and crucial molecular targets for preventive therapy are identified . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"neurobiology",
"of",
"disease",
"and",
"regeneration",
"animal",
"genetics",
"gene",
"regulation",
"neuroscience",
"animal",
"models",
"histology",
"behavioral",
"neuroscience",
"model",
"organisms",
"molecular",
"genetics",
"gene",
"expression",
"biology",
"mouse",
"genotypes",
"phenotypes",
"protein",
"translation",
"heredity",
"genetics",
"genetics",
"of",
"disease",
"genetics",
"and",
"genomics"
] | 2012 | ATXN2-CAG42 Sequesters PABPC1 into Insolubility and Induces FBXW8 in Cerebellum of Old Ataxic Knock-In Mice |
The response behaviors in many two-alternative choice tasks are well described by so-called sequential sampling models . In these models , the evidence for each one of the two alternatives accumulates over time until it reaches a threshold , at which point a response is made . At the neurophysiological level , single neuron data recorded while monkeys are engaged in two-alternative choice tasks are well described by winner-take-all network models in which the two choices are represented in the firing rates of separate populations of neurons . Here , we show that such nonlinear network models can generally be reduced to a one-dimensional nonlinear diffusion equation , which bears functional resemblance to standard sequential sampling models of behavior . This reduction gives the functional dependence of performance and reaction-times on external inputs in the original system , irrespective of the system details . What is more , the nonlinear diffusion equation can provide excellent fits to behavioral data from two-choice decision making tasks by varying these external inputs . This suggests that changes in behavior under various experimental conditions , e . g . changes in stimulus coherence or response deadline , are driven by internal modulation of afferent inputs to putative decision making circuits in the brain . For certain model systems one can analytically derive the nonlinear diffusion equation , thereby mapping the original system parameters onto the diffusion equation coefficients . Here , we illustrate this with three model systems including coupled rate equations and a network of spiking neurons .
The response behavior in two-choice decision making tasks is well described by so-called sequential sampling models [6] , of which Ratcliff's Diffusion model [7] , [8] is a particularly successful variant . According to the Diffusion model there is a decision variable X , the evolution in time of which is given by the linear diffusion equation ( 1 ) where ξ is zero-mean Gaussian white noise with unit variance , σ is the noise strength and ηL is a constant drift term . On each trial , X evolves from an initial condition until it reaches one of two fixed thresholds corresponding to the two decisions . The value of ηL depends on the strength of the task-relevant information in the stimulus and is typically taken to be constant within a trial but allowed to vary between trials according to some distribution . The Diffusion model can account for many of the observed phenomena in reaction time tasks ( e . g . [6] , [9] ) . In particular , making the task easier corresponds to increasing the value of ηL which leads to faster and more accurate decisions , e . g . [8] . The ability of a subject to trade speed for accuracy can be accounted for by changing the boundaries: by moving them closer to the starting points decisions become faster but less accurate [10] . Longer reaction times on error trials can be accounted for by introducing between-trial variability in the drift term , whereas shorter reaction times on error trials can be accounted for by considering a distribution of initial conditions [9] . The Diffusion model can be related to other behavioral models of decision making . It can be conceived of as a continuous-time version of random walk models ( e . g . , [11] ) . Usher and McClelland demonstrated how a two-dimensional connectionist model , with piecewise linear activation functions , can be reduced to a one dimensional diffusion equation [12] . The general form of the resulting diffusion equation differs from Equation 1 in that it includes a term linear in the decision variable and hence the resulting diffusion process is the so-called Ornstein-Uhlenbeck process . More recently Brown , Bogacz and co-workers have demonstrated that the linear diffusion equation can be derived from a range of different linear and piecewise linear connectionist and related models [13] , [14] . Recently , neuroscientists have begun to investigate the single-cell neurophysiology of the decision making process in two-choice tasks ( e . g . , [15]–[19] ) . In many of the tasks used by neuroscientists the subject is presented a visual stimulus and the behavioral response is indicated by a rapid eye-movement ( saccade ) to one of two pre-specified targets . Decision making related neuronal activity in these tasks has been described in a number of brain areas that are known to be involved in the planning and control of eye-movements: lateral intraparietal area ( LIP ) [20] , [21] , dorsolateral prefrontal cortex [22] , the frontal eye fields ( FEF ) [22] , and the superior colliculi [23] , [24] . There are several key features of the neuronal activity observed in these experiments that are important for our work: i ) The average firing rate of cells in these areas is correlated to the response behavior of the animals . This indicates that the firing rate “represents […] the information on which the developing decision is made” [21] . Further support for this interpretation comes from a study showing a direct correspondence between the time-evolution of trial-averaged firing rates in the superior colliculus and the dynamics of the decision variable in the Diffusion model [25] . ii ) The time-evolution of the trial-average firing rates is consistent with there being a competition between groups of cells associated with the two different decisions . The firing rate in the group associated with the correct decision shows , on average , an increase with time whereas the firing rate of the other group shows a decrease with time ( e . g . , [21] ) . Further evidence of a competition comes from microstimulation experiments [26] . iii ) The neurons in many of the involved areas show evidence of nonlinear interactions . In particular , many cells continue to fire at an elevated rate after the stimulus indicating where to move the eyes to is removed [20] . This so-called persistent activity can be accounted for by models of recurrent networks of spiking neurons [27] . Indeed , in a series of papers , Wang and co-workers have shown that biophysically motivated cortical network models of a two-choice decision making task can qualitatively replicate some salient aspects of both behavioral and neurophysiological data [28]–[31] . Such models posit two populations of recurrently coupled excitatory neurons each of which receives input proportional to the relative evidence in favor of the choice which it encodes . The populations compete through interneuron-mediated inhibition leading to winner-take-all behavior . On each ‘trial’ the state of the system evolves until the activity of one of the two populations exceeds a fixed threshold indicating a decision for that choice . In this model , making the task easier corresponds to increasing the input to one of the populations relative to the other [28] whereas the speed-accuracy trade-off has been accounted for by adjusting the threshold [30] . Assuming that the brain regions involved in the decision making process implement a winner-take-all strategy , as suggested by computational models , it remains unclear how this might lead to a response behavior best described by a one-dimensional diffusion processes . In other words , what is the relationship between the neuronal activity in putative decision making circuits and decision variables in behavioral models such as X in Equation 1 [32] ? Recent theoretical work exploring the dynamics of winner-take-all models for decision-making has shown that several models can , in fact , be reduced to a one-dimensional diffusion process provided that the models themselves are linear . Usher and McClelland [12] studied a two variable connectionist model with inhibitory cross-coupling and linear-threshold activation functions . While the thresholding is nonlinear and leads to bistable behavior , the reduction to a one-dimensional diffusion process is possible only in the region where the argument of the linear-threshold function is the same for both variables and the dynamics , therefore , are linear . Brown et al . [13] study both linear and piecewise linear systems , while Bogacz et al . [14] study the relationships between a number of linear connectionist models and show under which conditions these models can be formally reduced to a one-dimensional linear diffusion equation . It remains however , unclear how the dynamics of such linear systems might be related to that of more biologically realistic neural models , which exhibit strong nonlinearities . A first step towards resolving this issue was taken by Wong and Wang [29] in which they derive a reduced system of coupled nonlinear equations from a full spiking network via a semi-analytical approach . They then show that the linearization of the reduced system in the unbiased case can be reduced to a one-dimensional diffusion equation at the point where the spontaneous state destabilizes ( [29] , Text S1 ) . However , we may ask if the notion of a linear diffusion process is still valid once one takes into account nonlinear effects present in the system . Here we show how one can go beyond linearizations to take into account nonlinear effects in neural winner-take-all models . In particular , we will show how models of neuronal dynamics in two-choice decision making tasks can be formally reduced to a one-dimensional nonlinear diffusion equation . Instead of focusing on a particular model system we consider a generic model of the neuronal dynamics with two key features: nonlinearity and competition . One obvious advantage of using such a general framework is that the extent of validity of the reduction is potentially very large . Moreover , most detailed models of the neuronal underpinnings of decision making do include nonlinearity and competition as important ingredients ( e . g . , [28] , [33] , [34] ) . Unlike in linear systems , a proper reduction of the dynamics to one dimension in nonlinear winner-take-all models leads to a nonlinear diffusion equation . The nonlinear diffusion equation takes the form of a stochastically driven normal form for an imperfect pitchfork bifurcation . The nonlinear diffusion equation not only provides the correct qualitative description of the dynamics in neural winner-take-all models in general , but can also be derived from model systems if they are analytically tractable . This allows one , in the context of two-choice decision making , to determine how the coefficients of this diffusion equation functionally depend on neurobiologically meaningful quantities .
We begin on a fairly technical note in order to provide some sense of the generality of the result . We will then make use of , for illustrative purposes , a simple model which nonetheless retains some biophysical plausibility . While the method we use may seem complicated and the algebra is , in general , involved , the idea behind the reduction is simple . We take advantage of the dramatic reduction in dimensionality which occurs spontaneously in dynamical systems near a point where the qualitative behavior of the system changes , i . e . stationary states appear , disappear or change in nature . Such transition points or bifurcations , are ubiquitous in physical and biological systems , e . g . see [35]–[37] . Here we make use of the fact that winner-take-all models for two-choice decision making , irrespective of their dimensionality or complexity , generically undergo such a bifurcation when two new stationary states appear , corresponding to the two potential ‘winner/loser’ pairs . To emphasize the reduction in dimensionality we first consider a system of n nonlinear equations of the form ( 2 ) ( 3 ) ( 4 ) ( 5 ) where the Is are external inputs and xi represents the activity of the ith neuronal population , and represents the time derivative of x . Here x1 and x2 are populations whose activity correlates with the two possible developing choices while the remaining populations are non-selective given the particular task . We note that for I1 = I2 the equations are invariant under the transformation ( x1 , x2 ) → ( x2 , x1 ) , a property known as reflection symmetry . We see from this symmetry that the existence of the fixed point ( x1 , x2 , … ) = ( xhigh , xlow , … ) implies the existence of the fixed point ( x1 , x2 , … ) = ( xlow , xhigh , … ) . That is , if there is state in which population 1 exhibits a high level of activity and population 2 a low level of activity , then we are assured the existence of the opposite state . If these are the only possible stable states at long times then Equations 2–5 constitute a so-called ‘winner-take-all’ system . Since we want the system to behave in a winner-take-all fashion only when provided with sufficient input , we furthermore assume the existence of a fixed point , representing the spontaneous state . The derivation of the correct one-dimensional reduction of this system begins with the evaluation of the linear stability of this fixed point . We thus consider small perturbations of this state with the ansatz where xSS are the steady state values and perturbations with growth rate λ have the form . Plugging this ansatz into Equations 2–5 yields ( 6 ) where α is the derivative of f with respect to x1 in Equation 2 and x2 in Equation 3 , β is the derivative of f with respect to x2 in Equation 2 and x1 in Equation 3 , the γs are derivatives with respect to x1 and x2 and all derivatives are evaluated at the fixed point . It is clear that for α = β , which will occur only for a special parameter set , the first two rows cease to be linearly independent implying a zero eigenvalue with eigenvector xcr = ( 1 , −1 , 0 , … , 0 ) which corresponds to a mode for which either x1 or x2 increases while the other decreases , i . e . the winner-take-all dynamics we are interested in . If we wish our system to exhibit winner-take-all behavior then it must also be that the real part of the remaining n−1 eigenvalues is negative to avoid unwanted instabilities unrelated to the dynamics of interest and sufficiently distant from zero . This implies that perturbations along the eigenvector corresponding to the ‘winner-take-all’ instability neither decay nor grow linearly , while perturbations in any other direction quickly decay to zero . This is precisely the scenario in which a reduction to a one-dimensional dynamics is appropriate , as noted elsewhere , e . g [14] , [29] . Specifically , the dynamics along n−1 of the n dimensions will rapidly converge from an initial state to a one-dimensional manifold along which the dynamics are slow . This separation of time-scales , where the time-scales are inversely related to the eigenvalues of the linearized system , is what gives us the reduction in dimensionality . We now wish to derive an equation for the dynamics of the ‘winner-take-all’ instability . To do so we express the dynamical variables as x = xSS+xcrY ( T ) +--- , where Y represents the slow dynamics along the critical eigenvector and T is a slow time scale . Note that the reflection symmetry of the system implies that the dynamics of Y should be invariant under the transformation Y→−Y since this switches the identity of x1 and x2 . We assume that the increase in input , I , common to both x1 and x2 leads to the developing decision in the winner-take-all system and is thus the bifurcation parameter . This means that the linear growth rate of the spontaneous state must be proportional to the difference between the presynaptic input and the value of the input at the bifurcation although with an unknown prefactor , i . e . μ ( I−Icr ) . The difference in inputs , I1−I2 , breaks the reflection symmetry thereby introducing a constant term which , to first approximation , must be proportional to that difference although with an unknown prefactor , i . e . . These two facts , coupled with the reflection symmetry , lead to the form of the equation describing the time evolution of Y ( 7 ) where I = Icr only when α = β identically , i . e . at point of instability , and ∂T is a time derivative with respect to the slow time T . Note that for I1−I2 the equation is invariant under Y→−Y as it should be ( indeed , Y3 is the lowest order nonlinearity which obeys reflection symmetry ) . The coefficients , μ and γ can be calculated analytically from Equations 2–5 . This line of argumentation can be made more exact mathematically , see Materials and Methods and the supporting material ( Text S1 ) for examples from several systems , and is a standard technique in nonlinear dynamics known as multiple-scale analysis , see e . g . , [38] , [39] . For more complex systems which exhibit winner-take-all behavior , Equation 7 still captures the qualitative dynamics of the system near the bifurcation in general , although it may not be possible to calculate the coefficients . In addition , we are interested in the case of stochastically driven dynamics , which will lead , to leading order , to an additive noise term whose amplitude can also be calculated analytically for Equations 2–5 , see Materials and Methods and supporting material ( Text S1 ) for details . Finally , we arrive at the equation ( 8 ) where and the sign of γ determines the sign of the cubic term . We note here that although we have not derived Equation 8 from any particular system ( and thus we do not know the functional dependence of η , μ , and σ on relevant physiological parameters ) , we nonetheless do know the leading order dependence on changes in external inputs to the two ‘competing’ populations . Thus the constant drift term is linear proportional to differences in these inputs while the linear term is proportional to the common input to both populations and is exactly equal to zero at the critical value . We also note that the evolution of X in Equation 8 can be thought of as the motion of an noise-driven , overdamped particle in a potential , or ( 9 ) ( 10 ) The use of an analogy to an ‘energy landscape’ as an intuitive explanation for the dynamics in neural winner-take-all models is not new , e . g . [29] . However , here we have gone beyond analogy to show the actual form of the potential . The framework we chose above for illustration , Equations 2–5 , is a system of coupled nonlinear ordinary differential equations . However , the derivation of the nonlinear diffusion equation Equation 8 is not contingent on the original system having this particular form . For this reason we have chosen to illustrate the derivation of the nonlinear diffusion equation from three distinct model systems . Below we study a system of three coupled rate equations . In the supporting material ( Text S1 ) we consider a system of three populations of integrate-and-fire neurons . Following the work of Brunel and Hakim [40] we recast the network as a system of coupled partial differential equations ( Fokker-Planck equations ) describing the evolution of the probability densities for the voltages . The nonlinear diffusion equation can then be derived from the system of partial differential equations . Finally , in the supporting material ( Text S1 ) we also include a derivation from two coupled rate equations which Wong and Wang derived via semi-analytical arguments from a full spiking network [29] . We consider a simple model describing the activity of two excitatory populations of neurons which compete via a population of inhibitory interneurons . The equations are ( 11 ) ( 12 ) ( 13 ) where rI , r1 and r2 are the activity of the inhibitory and two excitatory populations respectively . The input to each population consists of a combination of recurrent and external inputs . For each excitatory population there is a recurrent excitatory coupling of strength s , an inhibition term with strength c and an input made up of a common and population specific parts , I and Ii where i = 1 , 2 . The inhibitory population receives excitatory drive from both populations with a strength g and input II and we have neglected any self-inhibition term . The Φs are nonlinear transformations of the input . Fluctuations in the activity variables are expressed via unit variance Gaussian white noise terms ξ ( t ) with strength σE and σI for the excitatory and inhibitory populations respectively . Additionally , we have normalized time by the time constant of the excitatory populations and τ thus represents the ratio of the inhibitory to the excitatory time constant . For this system a ‘winner-take-all’ instability occurs for sΦ′ ( I = Icr ) = 1 . Following the general framework outlined above one carries out a multiple-scale analysis ( see Materials and Methods ) to arrive at Equation 8 with ( 14 ) ( 15 ) ( 16 ) ( 17 ) where all derivatives of Φ are evaluated at I = Icr . Note that here the slow dynamics at the point of instability ( bifurcation ) are not dominated by the time constant τ ( nor by 1 ) but rather by the near-zero eigenvalue of the critical mode . Thus extremely slow dynamics can thus be achieved in the vicinity of the bifurcation even in the absence of an intrinsic slow time constant such as that due to the activation of recurrent NMDA receptors , e . g . [28] . The presence of slow intrinsic time constants is , however , beneficial in eliminating oscillations and for obtaining realistic firing rates for working memory states , e . g . [28] , [41] . Generically , two qualitatively different scenarios for winner-take-all dynamics can occur in nonlinear systems . The explicit dependence of the coefficients in Equation 8 on the inputs to the two populations allows us to directly relate modulations in these inputs to changes in reaction-times and performance . In doing so we will make use of the formulation of a nonlinear diffusion equation as the motion of a particle in a potential , Equations 9–10 , see Figure 2 . An increase in the difference of the two external inputs tilts the potential in favor of the population with the greatest input , an effect also seen in linear , connectionist models , e . g . [12] . Modulations of the input common to both populations affect the curvature of the potential . For common inputs below the bifurcation , the potential exhibits a dimple , reflecting an attracting spontaneous state , while above the bifurcation the spontaneous state is repelling . This modulation of the potential via changes in the common input is a consequence of the nonlinearity of the system . In linear systems , changes in the mean input , given a fixed threshold , shift the position of the effective threshold for the decision making process , e . g . [14] . Below , we discuss these effects in greater detail , making use of exact expressions for reaction times RT ( X0 ) and performance P ( X0 ) as a function of the initial condition X0 . See supporting material ( Text S1 ) for the expressions . When both populations receive the same mean input , i . e . Δν = 0 , the energy function , Equation 10 , is symmetric , leading to an equal probability of escape through either boundary , i . e . performance P ( 0 ) = 0 . 5 , see Figure 2A green . If one population receives more input than the other , i . e . its activity encodes the ‘correct’ choice , Δν≠0 and the probability of making the corresponding choice will be greater than chance , P ( 0 ) >0 . 5 . This is reflected in the asymmetry of the energy function , which is now tilted towards the correct choice , see Figure 2A black . Reaction-times for the correct choice decrease monotonically with increasing Δν . Reaction-times are , in general , different for the error choice with respect to the correct choice for a fixed value of Δν and can exhibit non-monotonic dependence on Δν , see supporting material ( Text S1 ) . Mean reaction-times for error trials can , in fact , be slower or faster than those for correct trials depending on the value of the common input . Performance increases with increasing Δν , owing to a more pronounced asymmetry in the energy . Indeed , it can be shown analytically that , see supporting material ( Text S1 ) . An analogous effect is obtained in the linear diffusion equation , Equation 1 , by changing the constant drift term . Changes in the input common to both populations affect the quadratic term in E ( X ) through . For the spontaneous state is stable , reflected in the local minimum of the energy shown in Figure 2B red . As the common inputs are increased , increases , destabilizing the spontaneous state and thus converting the local minimum to a local maximum for , Figure 2B blue . For identically equal to zero , the spontaneous state is marginally stable , resulting in an energy function with zero curvature locally , Figure 2B green . Indeed , in this regime the dynamics in the vicinity of the spontaneous state behave similarly to those seen in the linear diffusion equation Equation 1 , whose energy function is given by E ( X ) = −ηLX with absorbing boundaries . Since reaction-times are given by the time it takes for the system to escape from the spontaneous state , it is clear that reaction-times decrease with increasing as the spontaneous state changes from attracting ( local minimum ) to repelling ( local maximum ) . In fact , it can be shown analytically that for Δν = 0 , showing that reaction-times strictly decrease with increasing common input . Numerical investigation show that this holds also for Δν≠0 ( not shown ) . Furthermore , it can be shown analytically that for any value of Δν , see supporting material ( Text S1 ) . The fact that both reaction-times and performance decrease monotonically with increasing common input suggests a novel mechanism to explain the physiological underpinnings of the speed-accuracy trade-off . That is , Equation 8 predicts that increases in common input to the two populations will lead to faster reactions and poorer performance while decreases in input will lead to slower reactions and better performance . We note that increasing the noise amplitude σ in Equation 8 leads to decreasing performance . Increasing noise amplitude also tends to reduce reaction-times given initial conditions in the vicinity of the spontaneous state . To reiterate , Figure 2 shows how the energy landscape and hence the system dynamics in two-population winner-take-all networks is affected by changes in afferent inputs alone . This is important since the dependence on the input holds for all models irrespective of the details , while changes in the coefficients η , μ , and σ imply changes in single-cell and network properties specific to the model chosen . We now show that such changes in input are sufficient to describe behavioral data in two-choice decision making tasks . Specifically , we consider data from two separate studies using the so-called random moving dots task , namely from Roitman and Shadlen [21] and Palmer et al . [10] . The coherence of the stimulus in these experiments is defined as the fraction of dots moving in one of two possible directions , the remaining dots moving randomly . The subject must indicate the direction of the coherent motion with a saccade . We choose this particular task as behavior and electrophysiologal activity have been well characterized [10] , [20] , [21] , and biologically motivated two-population winner-take-all models have been evoked to describe the decision making process [28]–[30] . In particular , it has been shown that the stimulus coherence is encoded approximately linearly in the firing rates of direction-selective cells in area MT [44] . Evidence suggests that output from MT cells then drives neurons in area LIP whose trial-averaged activity is consistent both with the notion of a linear integrator ( ramping activity with increasing slope for increasing coherence ) and with that of competing populations of neurons ( activity ramps up or down depending on whether the receptive field is in the preferred or anti-preferred direction of motion respectively ) [20] , [21] . In light of these experimental observations , it is reasonable to assume that the difference in inputs from MT cells to the putative neuronal populations in LIP which encode the two possible directions , increases linearly with increasing coherence . Therefore , we assume a linear dependence of Δv on the stimulus coherence in Equation 8 . Doing so provides an excellent fit to behavioral data , capturing performance as well as both correct and error reaction times , without having to vary any additional parameters , see Figures 3 and 4 , solid line fit to symbols with error bars . Note that the difference in reaction-times for correct and error trials comes about due to the nonlinearity of the energy function which sits along the slow manifold ( mean correct and error reaction times are identical given a linear energy function ) . This is in contrast to the mechanism evoked in [29] to explain longer error reaction times using a two-component system of rate equations . There they argued that asymmetries in the phase plane lead to trajectories for error trials which stayed closer to the stable manifold as they approached the unstable manifold . The trajectories therefore came closer to the saddle-point leading to long residence times before escaping . This mechanism relies on the full two-dimensionality of the system coming into play . It is likely that the mechanism we describe here is dominant near the bifurcation , while far from the bifurcation the full dimensionality of the system being studied must be taken into account in order to explain longer error reaction times . We now show that the trade-off between speed and accuracy , commonly observed in reaction-time experiments [4] , can be explained through changes in the common input to the two populations . The data shown in Figure 4 are from three sets of experiments in which human subjects are told to respond within 0 . 5 , 1 and 2 seconds , and are shown in blue , red and black symbols respectively [10] . These data clearly exhibit precisely the speed-accuracy trade-off . As mentioned in the previous section , changes in may potentially capture this effect . Indeed , decreasing the input common to both populations increases reaction-times and performance , while increasing the input has the opposite effects . As seen in Figure 4 , changes in the input common to both populations , i . e . do in fact capture the speed-accuracy trade-off . Since Equation 8 can be derived analytically from more complex model systems , we can map the values of the coefficients obtained from fits to behavioral data back to more physiologically meaningful parameters . An example of this is shown in Figures 3 and 4 ( red symbols ) where we have conducted simulations of the rate equations Equations 11–13 with parameter values chosen to match the coefficients from the fit , using Equations 14–17 . Thus one can trivially fit higher dimensional models to data once Equation 8 has been derived . The fits of the nonlinear diffusion equation Equation 8 to behavioral data suggest not only that the putative decision making circuit behaves in a way consistent with a winner-take-all framework , but that changes in inputs to this circuit alone are sufficient to account for performance and mean reaction-times . In addition , the best fits to the data were found for |ηΔν| , , i . e . in the vicinity of the bifurcation . This provides an a posteriori validation of the use of Equation 8 to describe these data since it represents , after all , a reduction of the dynamics near the bifurcation . Moreover , it is precisely in this regime that the dynamics of Equation 8 most closely resembles that of the linear diffusion equation .
One-dimensional diffusion equations have long been used to model behavior in two-choice reaction-time tasks . Recently , researchers discovered that the trial-averaged single-unit activity recorded in areas of the brain which are implicated in generating this behavior closely resemble the dynamics of a linear diffusion process [21] . This suggests a correspondence between the neural activity in these areas and the decision making variable X in the linear diffusion equation . However , it remained unclear how the cortical activity might actually conspire to generate such a linear diffusion . On the other hand , it was soon demonstrated that some aspects of the neural activity could be captured in biophysically motivated winner-take-all network models [28]–[30] . Here we have shown , through the use of standard tools from nonlinear dynamics theory , that the dynamics in winner-take-all models relevant for two-choice decision making can be captured in a one-dimensional nonlinear diffusion equation , Equation 8 . This suggests that the cortical circuits involved in decision making generically generate an effective nonlinear diffusion which in a limited parameter regime leads to behavior very similar to that predicted by the linear diffusion equation . The dependence of the coefficients in Equation 8 on external inputs is explicit and independent of the details of the underlying model . This suggests that the functional dependence of behavioral measures in two-choice decision making on changes in inputs is universal . In particular , we predict that modulations of the input common to both populations can account for the speed-accuracy trade-off . This mechanism differs from that evoked by others previously , which consists of varying the threshold for detection of the decision ( e . g . a higher threshold increases reaction times and increases performance ) , [6] , [30] . The novel mechanism proposed here of speed-accuracy trade-off through modulations in the mean input predicts that pre-stimulus activity in LIP should be higher , on average , when the subject must respond more rapidly . Support for this comes from the observation that the baseline neuronal activity in monkeys varies in a task-dependent manner , see Figure 16 from [20] , a phenomenon which has been interpreted as anticipatory activity . Indeed , increases in the baseline activity were found to correlate with more rapidly evolving post-stimulus activity . Equation 8 now provides us with an explanation for the functional role of this activity . This phenomenon could be further confirmed through comparison of the relative changes in the BOLD signal in fMRI studies of activity in brain areas in humans homologous to LIP during the pre-stimulus period in a task where the speed-accuracy trade-off is observed behaviorally . While Equation 8 appears similar in form to other diffusion models which have been used to describe behavior in two-choice decision making [9] , [12]–[14] , [45] , it is important to distinguish between their very distinct mathematical pedigrees . In particular , we have not evoked the nonlinear diffusion equation as a phenomenological model of behavior for two-choice decision making . Rather , it represents the correct asymptotic description of the dynamics in nonlinear winner-take-all models near the bifurcation to winner-take-all behavior . This observation has two consequences . Firstly , in as far as nonlinear winner-take-all models can successfully reproduce some qualitative features of the neuronal activity in brain areas implicated in the decision making process for two-choice decision making [28] , i . e . LIP , the nonlinear diffusion equation also provides an approximate description of this activity . Secondly , if an actual nonlinear winner-take-all process is at work in the brain during such tasks , then this process will behave as an approximately one-dimensional diffusion process in the vicinity of the bifurcation to winner-take-all behavior . This process is described by the nonlinear diffusion equation Equation 8 . Note also that the effective reduction in dimension of the dynamics in nonlinear systems in general only occurs at bifurcations . Thus nonlinear normal forms for bifurcations such as Equation 8 represent the only proper one-dimensional reduction of such a system . As in the linear diffusion equations , bias in external inputs in the nonlinear diffusion equation appears to leading order as a constant drift term . In contrast , while reductions of linear connectionist models to the linear diffusion equation lead to a linear ( Ornstein-Uhlenbeck ) term proportional to the difference between intrinsic ‘leak’ and the effective cross inhibition , this is not the case in nonlinear systems . Rather , this term reflects the linear growth rate of the spontaneous state which , given that the input is the bifurcation parameter , is simply proportional to the distance of the common external input from the critical value at the bifurcation . Thus this term varies with modulations of the external input , unlike in the linear case . Finally , the cubic nonlinearity , which is the lowest order nonlinearity consistent with the reflection symmetry of the original system , leads to an inverted-U potential . This drives the activity to infinity in finite time , reflecting the escape from the spontaneous state to the ‘decision’ state . As illustrated in Figure 1 , this renders the measurement of reaction-times and performance insensitive to the exact placement of a threshold as long as it is high enough . Setting relatively high thresholds therefore effectively eliminates one free parameter from the model , namely the threshold placement . Nonetheless , one could set low thresholds in the nonlinear system , i . e . very close to the spontaneous state [30] . It has been hypothesized that the threshold for detection of a decision in the brain may be set by downstream areas including superior colliculus [30] or the basal ganglia [46] As it turns out , Equation 8 can account for behavioral data for the random moving dot task in monkeys and humans , c . f . Figures 3 and 4 . As such Equation 8 seems to provide a correct description of both the neuronal activity and the behavior in this task , thereby linking the two . This , however , in no way contradicts the success of connectionist and linear diffusion models in fitting behavioral data . Indeed , a comparison of the nonlinear diffusion equation and the linear one , Equation 1 , shows approximately equally good fits for correct reaction-times and performance for the data in Figures 3 and 4 , see supporting material ( Text S1 ) . On the other hand error reaction-times in Figure 3 , which are longer than correct ones , cannot be fit by the linear diffusion model unless variability in the initial condition and drift term across trials is introduced [9] . They are , however , correctly captured by the nonlinear diffusion equation . We note , furthermore , that several groups have derived reduced models for two-choice decision making . Wong and Wang performed a heuristic reduction of a spiking network model to a system of two coupled rate equations [29] , and showed that it gave similar qualitative behavior . As a canonical model , Equation 8 qualitatively captures the dynamics of both the network and the rate models , also see fit in supporting material ( Text S1 ) . We note , however , that far from the bifurcation the full dimensionality of the system being studied will come into play and the dynamics will not be captured by Equation 8 . Much of the phenomenology in [29] appears to occur in this regime . Once this is the case , the dynamics may depend crucially on the details and dimensionality of the system and , if so , cannot be generalized . Wong et al . have recently used their reduced model to explain the experimentally observed violation of time-shift invariance in the behavior of monkeys doing the random moving dot task [47] , lending further support for the nonlinear , attractor network framework for LIP activity [31] . They also note that the inclusion of target inputs , which more faithfully reproduces the experimental paradigm , ‘renders the model behavior closer to a one-dimensional model in the decision process’ [31] . Interestingly , the presence of the unstable cubic term in the 1D nonlinear diffusion equation Equation 8 should lead to the experimentally observed violation of time-shift invariance for which perturbations arriving later in time have a lesser effect due to the nonlinear acceleration away from the spontaneous state . This remains to be tested quantitatively . Usher and McClelland derived a one-dimensional diffusion equation equivalent to an Ornstein-Uhlenbeck process from a neurobiologically motivated system of two coupled , threshold-linear equations [12] . This and other similar systems of linear equations were studied by Bogacz , Brown and collaborators [13] , [14] . The linearity of the system in these studies allowed for an in-depth analytical characterization of the dynamics . Indeed , it has been argued that neurobiologically motivated models might , within certain parameter regimes , be reducible to an equivalent linear diffusion equation [14] . However , as we have shown here , if the underlying winner-take-all system exhibits any generic nonlinearities , as seems to be the case in neural systems , the correct dynamics are given by Equation 8 . Soltani and Wang [48] and Fusi et al . [49] have both investigated how synaptic plasticity might shape the response in winner-take-all decision making circuits . Soltani and Wang introduced a reward-dependent stochastic Hebbian rule for updated synaptic strengths which successfully reproduces the so-called ‘matching behavior’ while Fusi et al . have presented a model of flexible sensorimotor mapping in which reward-dependent synaptic plasticity shapes the output of a winner-take-all decision making circuit . In both cases , the performance depends on the difference in the fraction of potentiated synapses between the two populations Δc , i . e . the symmetry breaking occurs due to plastic changes in synaptic strength . In the context of Equation 8 this would lead to an additional term which is functionally equivalent to the symmetry-breaking term proportional to the difference in inputs . The effect of synaptic plasticity in two-choice decision making could therefore be studied by means of Equation 8 coupled with an appropriate learning rule . The reduction to Equation 8 is strictly valid only in the immediate vicinity of the bifurcation . For this reason it might be argued that the current scenario is tantamount to fine-tuning and may not be biologically relevant . Three facts indicate this is not the case . ( I ) As we have shown here Equation 8 can be rigorously derived from model systems and can provide a quantitative match even away from the bifurcation . ( II ) Equation 8 can be fit to behavioral data , previously published model networks [28] and models in regimes far from the bifurcation where a quantitative match is no longer found . It thus provides a correct qualitative description of the dynamics . Furthermore these fits are made by varying physiologically meaningful parameters in ways that are either consistent with experimental findings or which lead to experimentally testable predictions . ( III ) Lastly , a large literature exists showing that human behavior in 2-choice decision making is well-described by one-dimensional sequential sampling models . A deep question is how such low-dimensional dynamics might arise from high-dimensional neuronal dynamics . We believe the most parsimonious explanation is that the neuronal circuits involved operate near the low dimensional manifold which arises naturally within a certain parameter range , i . e . near the bifurcation .
Here we derive the nonlinear diffusion equation ( noise driven amplitude equation for an imperfect pitchfork bifurcation ) from Equations 11–13 . We first study the linear stability of the spontaneous fixed point , analogously to Equation 6 and then extend this analysis to take into account nonlinear effects in a so-called weakly nonlinear analysis using a multiple-scales approach . We assume that IA = IB = I and consider an ansatz of the form ( rA , rB , rI ) = ( R , R , RI ) + ( ΔrA , δrB , δrI ) eλt , where R = Φ ( sR−cRI+I ) and RI = ΦI ( 2gR+II ) . This leads to an eigenvalue problem of the form ( 18 ) where the derivatives of the transfer functions are evaluated at the fixed point . The eigenvalue corresponding to the eigenvector ( 1 , −1 , 0 ) is equal to zero for . We expand the input current and the rates around the steady instability found above . We take ( 19 ) ( 20 ) ( 21 ) where ε and are small parameters which measure the distance from the bifurcation and the difference in inputs to the two excitatory populations respectively . Near the bifurcation , the mode corresponding to the critical eigenvector Y ( T ) evolves on the slow time scale T = ε2t . The expansions given above are plugged into Equations 11–13 and terms are collected order by order . We assume that , i . e . weak symmetry breaking . The scaling of input currents in ε is dictated by the reflection symmetry of the original system , i . e . we expect a pitchfork bifurcation . Were we to not use the knowledge of this symmetry , a more general expansion of the currents , including all orders of ε could be used , leading to the same result . That is , we would find for example that the term proportional to ε is identically equal to zero , etc . We solve for the performance and reaction time in Equation 8 ( solid lines in Figures 3 and 4 ) by numerically evaluating RT ( 0 ) and P ( 0 ) , Equations S2 and S4 , using Romberg integration [50] , with limits of integration of ±0 . 21 and ±0 . 19 for Figures 3 and 4 respectively . The value of RT and P are relatively insensitive to increases in the limits of integration , related to the fact that in Equation 8 , X approaches ±∞ in finite time . We have also fit the data from Figures 3 and 4 using direct numerical simulation of Equation 8 with a threshold of ±1 , obtaining results for RT which vary by no more than a constant shift of 10 ms . Fits in Figures 3 and 4 are made by eye . Once the fits have been made using the nonlinear diffusion equation , we must choose parameters in the rate equations which give the proper values for the coefficients , using the expressions Equations 14–17 . Various parameter combinations are possible , indicative of the reduction in dimensionality of the system and a potential mechanism for robustness in functionality . For the simulations in Figure 3 we took with α = 1 . 5 , β = 2 . 5 and x0 = 1 , s = 1 . 9 , c = 1 , g = 1 , II = 0 . 2 , I = 0 . 3695 , σE = σI = 0 . 001634 I1−I2 = 2 . 168e−05×coherence . For the simulations for subject 1 in Figure 4 we took with α = 1 . 5 , β = 2 . 5 and x0 = 1 , s = 1 . 9 , c = 1 , g = 1 , II = 0 . 2 , I = 0 . 3675 , 0 . 3687 , 0 . 3742 , σE = σI = 0 . 001634 I1−I2 = 4 . 066e−05×coherence . For the simulations for subject 2 in Figure 4 we took with α = 1 . 5 , β = 2 . 5 and x0 = 1 , s = 1 . 9 , c = 1 , g = 1 , II = 0 . 2 , I = 0 . 3673 , 0 . 3684 , 0 . 3721 , σE = σI = 0 . 001634 I1−I2 = 4 . 228e−05×coherence . In all cases , a trial ends once one of the rates crosses a fixed threshold of 0 . 7 . Initial condition was rA = rB = 0 . 16 , rI = 0 . 35 where the values at the bifurcation are rA = rB = 0 . 253 and rI = 0 . 486 . Changing the initial condition did not alter the results significantly ( not shown ) . We conducted 10 , 000 runs for each value of the coherence . | The brain holds a central position in scientific theories of rational behavior . For example , brain activity is thought to stand in a causal relation to the decision making behavior observed in two-choice perceptual discrimination tasks . Although a lot is known about both the brain activity and the response behavior during these tasks , the relationships between the two are not fully understood . In particular , how can one relate the high-dimensional dynamic activity of the brain to the low-dimensional descriptions of response behavior such as performance and reaction-times ? Our approach to this question is to relate existing neurobiological models of brain activity to existing models of response behavior . In this paper we establish a formal link between standard , winner-take-all models of brain activity during two-choice tasks and a family of one-dimensional behavioral models known as diffusion models . Our analysis demonstrates a universal functional dependence between the external inputs to the neural populations in the neurobiological model on the one hand , and reaction times and performance in the one-dimensional model on the other . Importantly , we show that experimentally measured performance and reaction-times can be predicted through changes in these external inputs alone . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"neuroscience/cognitive",
"neuroscience",
"neuroscience/theoretical",
"neuroscience",
"computational",
"biology/computational",
"neuroscience"
] | 2008 | Neurobiological Models of Two-Choice Decision Making Can Be Reduced to a One-Dimensional Nonlinear Diffusion Equation |
Schistosomiasis is a water-based disease that is believed to affect over 200 million people with an estimated 97% of the infections concentrated in Africa . However , these statistics are largely based on population re-adjusted data originally published by Utroska and colleagues more than 20 years ago . Hence , these estimates are outdated due to large-scale preventive chemotherapy programs , improved sanitation , water resources development and management , among other reasons . For planning , coordination , and evaluation of control activities , it is essential to possess reliable schistosomiasis prevalence maps . We analyzed survey data compiled on a newly established open-access global neglected tropical diseases database ( i ) to create smooth empirical prevalence maps for Schistosoma mansoni and S . haematobium for individuals aged ≤20 years in West Africa , including Cameroon , and ( ii ) to derive country-specific prevalence estimates . We used Bayesian geostatistical models based on environmental predictors to take into account potential clustering due to common spatially structured exposures . Prediction at unobserved locations was facilitated by joint kriging . Our models revealed that 50 . 8 million individuals aged ≤20 years in West Africa are infected with either S . mansoni , or S . haematobium , or both species concurrently . The country prevalence estimates ranged between 0 . 5% ( The Gambia ) and 37 . 1% ( Liberia ) for S . mansoni , and between 17 . 6% ( The Gambia ) and 51 . 6% ( Sierra Leone ) for S . haematobium . We observed that the combined prevalence for both schistosome species is two-fold lower in Gambia than previously reported , while we found an almost two-fold higher estimate for Liberia ( 58 . 3% ) than reported before ( 30 . 0% ) . Our predictions are likely to overestimate overall country prevalence , since modeling was based on children and adolescents up to the age of 20 years who are at highest risk of infection . We present the first empirical estimates for S . mansoni and S . haematobium prevalence at high spatial resolution throughout West Africa . Our prediction maps allow prioritizing of interventions in a spatially explicit manner , and will be useful for monitoring and evaluation of schistosomiasis control programs .
Schistosomiasis is a water-based disease caused by trematodes of the genus Schistosoma . The five schistosome species that are known to infect humans are Schistosoma mansoni , S . haematobium , S . intercalatum , S . mekongi , and S . japonicum . School-aged children are at highest risk of infection and are the main target group for interventions [1] . Despite successful efforts to control schistosomiasis in different parts of the world , more than 200 million individuals are still estimated to be infected and the annual global burden due to schistosomiasis might exceed 4 . 5 million disability-adjusted life years ( DALYs ) lost [1]–[3] . A substantial amount of this burden is concentrated in West Africa , including Cameroon . Indeed , 72 million infections are thought to occur in this part of the world [4] . However , the current statistics , as presented by Chitsulo et al . ( 2000 ) [4] , Steinmann et al . ( 2006 ) [2] , and Utzinger et al . ( 2009 ) [5] , are largely based on population re-adjusted data originally published by Utroska and colleagues in the late 1980s [6] . Hence , the estimates are likely to be outdated due to , among other reasons , large-scale preventive chemotherapy campaigns , improved sanitation , water resources development and management , and socio-economic development . Recently , donors have provided new funds to control the so-called neglected tropical diseases ( NTDs ) , including schistosomiasis . For cost-effective planning and evaluation of control activities , it is essential to have reliable baseline maps of the geographical distribution of at-risk population and disease burden . Early schistosomiasis mapping efforts have been based on climatic suitability thresholds [7] , [8] . These maps are not reliable because they are not based on disease data . Apart from a few studies [9]–[12] , empirical maps of disease distribution over large areas are not available since there is a paucity of contemporary large-scale survey data . The first comprehensive compilation of historical schistosomiasis prevalence surveys at a global scale was carried out by Doumenge et al . in the mid-1980s [13] . More recent collections are available by Brooker et al . ( 2010 ) [14] for soil-transmitted helminthiasis and schistosomiasis , but data access is limited . The European Union ( EU ) -funded CONTRAST project initiated the development of an open-access global NTD database , which is updated in real time ( GNTD database; http://www . gntd . org ) [15] . A key objective of CONTRAST is to employ this database for large-scale schistosomiasis prevalence mapping and prediction in sub-Saharan Africa for the spatial refinement of control interventions and the cost-effective allocation of resources . Geographical locations in close proximity share common exposures which influence the disease outcome similarly . The geographical information of the survey locations in the GNTD database allows taking into account the potential spatial correlation and therefore creation of more realistic models . Standard statistical modeling approaches assume independence between locations [16] . Ignoring potential spatial correlation in neighboring areas due to common exposures could result in incorrect model estimates [17] . Geostatistical models take into account spatial clustering by introducing location-specific random effect parameters in the covariance matrix by a function of distance between locations [16] . Such models typically contain large numbers of parameters and cannot be estimated by the commonly used maximum likelihood approaches [18] . Bayesian model formulations enable model fit via Markov chain Monte Carlo ( MCMC ) simulations [16] . Bayesian geostatistical models have been applied in mapping schistosomiasis at different spatial scales , for example by Raso et al . ( 2005 ) [19] in the region of Man , western Côte d'Ivoire , and Clements et al . ( 2008 ) [9] in Mali , Niger , and Burkina Faso . Brooker et al . ( 2010 ) [14] developed a global predictive map highlighting those areas where preventive chemotherapy against schistosomiasis and soil-transmitted helminthiasis are warrant . However , to our knowledge , there is neither a model-based S . haematobium nor a S . mansoni large-scale prevalence map and spatially explicit burden estimates for the whole West African region . In this paper , we developed Bayesian geostatistical models based on environmental and climatic risk factors to obtain reliable empirical schistosomiasis prevalence maps for individuals aged ≤20 years by analyzing the GNTD data for West Africa , including Cameroon . Prediction was based on joint kriging in order to summarize the results as population-adjusted country prevalence estimates . Emphasis was placed on the distribution of S . haematobium and S . mansoni . We neglected S . intercalatum due to low infection risks , especially outside Cameroon .
The GNTD database was used to obtain prevalence data on schistosomiasis . This database assembles general information about the type of publication , authors , and publication year , as well as study-specific information about survey population , survey period , Schistosoma species , diagnostic test employed , and the number of infected individuals among those examined , stratified by age and sex ( if available ) . Hospital studies , data on specific susceptible groups ( such as HIV positives ) , and post-intervention studies were not included in the database [15] . For this study , we analyzed all point-level data on settled populations in West Africa on either S . haematobium or S . mansoni: 4550 and 2611 survey locations , respectively . We excluded ( i ) surveys with missing geographical coordinates; ( ii ) missing numbers of individuals screened; ( iii ) surveys carried out before 1980; ( iv ) individuals aged >20 years; and ( v ) entries based on certain diagnostic techniques . With regard to the latter exclusion criteria , we rejected all non-direct diagnostic examination techniques , such as immunofluorescence tests , antigen detections or questionnaire data , and direct fecal smears that have very low diagnostic sensitivities ( overall , 4% of the data for S . mansoni and 0 . 1% for S . haematobium were excluded ) . Hence , the surveys included were mainly based on the Kato-Katz thick smear method ( S . mansoni ) and urine filtration or sedimentation ( S . haematobium ) . Sensitivity and specificity of the diagnostic techniques were not incorporated in the model due to usually unknown sampling effort ( e . g . , number of stool samples , number of slides examined under a microscope , etc . ) , which affect diagnostic accuracy . We assumed that the proportion of rejected diagnostic techniques among the data with missing information on the technique ( S . mansoni: 33 . 5% missing , S . haematobium: 20 . 6% missing ) is similar . Therefore , we considered the bias that would arise from ignoring the missing data as larger than the bias from potentially rejected diagnostic techniques among the missing data . A separate model validation on the reduced datasets confirmed that by including data with incomplete records the predictive ability increased compared to the model excluding this information ( results not presented ) . Climatic , environmental , and population data were obtained from different freely accessible remote sensing data sources , as summarized in Table 1 . Data on day and night temperature were extracted from land surface temperature ( LST ) data . The normalized difference vegetation index ( NDVI ) was used as a proxy for vegetation . Digitized maps on freshwater body sources ( e . g . , rivers , lakes , and wetlands ) in West Africa were acquired with the characteristic of being either perennial or temporary . Processing of the MODIS/Terra data was carried out using the ‘MODIS Reprojection Tool’ [20] and code implemented in Fortran 90 [21] to summarize the temporal changes by an overall yearly average based either on the mean ( NDVI , day and night LST ) or the mode ( land cover ) . Furthermore , the land cover categories , as defined by the International Geosphere-Biosphere Programme , were re-grouped into six categories as follows: ( i ) sparsely vegetated; ( ii ) deciduous forest and savanna; ( iii ) evergreen forest; ( iv ) cropland; ( v ) urban; and ( vi ) wet areas . Rainfall estimates were processed via the software IDIRSI 32 [22] . Yearly averaged rainfall was calculated as summary measure . Distance calculations to the nearest freshwater body source were done in ArcMap version 9 . 2 of the Environmental Systems Research Institute ( ESRI; Redlands , CA , USA ) [23] . A classification scheme of West Africa into ecological zones was obtained using a demo version of the Earth Resources Data Analysis System Imagine 9 . 3 software [24] . The datasets were subjected to an unsupervised classification , via the ‘Iterative Self-Organizing Data Analysis Technique’ ( ISODATA ) , to map areas of environmental clustering which were further summarized into five main classes based on between-class similarities . The resulting map matched existing classifications [25] and the classes can be interpreted as ( i ) desert/semi-desert; ( ii ) sahelian zone; ( iii ) savannah; ( iv ) forest; and ( v ) tropical rainforest . Population count data obtained from LandScan for 2008 were converted to 5×5 km spatial resolution and adjusted to 2010 using country-specific average annual rates of change for 2005–2010 provided by the United Nations ( UN ) [26] . Estimates for the percentage of individuals aged ≤20 years among the total population per country were extracted from the U . S . Census Bureau International Database [27] for the year 2010 . Population counts were linked to the percentage of children . The estimated number of infected individuals ≤20 years was calculated by combining a sample of the joint predictive posterior distribution of the disease prevalence predicted at pixel level with the population size of that age group within the pixel . The predictive posterior distribution of the number of infected individuals per country was estimated by summing up the pixel-samples and calculating summary statistics . The combined schistosomiasis prevalence ( infection with S . mansoni or S . haematobium or both ) was calculated on the assumption that the two infections are independent from each other , as Schistosoma spp . = S . mansoni+S . haematobium− ( S . mansoni * S . haematobium ) . Extraction of the remotely sensed data at the survey locations and at the prediction locations for the two databases was performed via a self written Fortran 90 code . The prediction surface for West Africa was built in ArcMap [23] with a spatial resolution of 0 . 05°×0 . 05° ( approximately 5×5 km ) resulting in approximately 220 , 000 pixels covering the study region . The data were displayed in ArcMap . For each Schistosoma species , bivariate logistic regressions were performed in STATA/IC 10 . 1 [28] in order to assess potential covariates in relation to the outcome ( the number of infected individuals over the number of individuals screened per location ) . Continuous covariates were categorized into four groups based on quartiles to account for potential non-linearity in the outcome-predictor relationship on the logit . The Bayesian information criterion ( BIC ) was employed to detect whether linear or categorized covariates on the logits have smaller BIC and therefore predict the outcome more accurately . We used the following covariates in both linear and categorical scales: altitude , day LST , night LST , rainfall , NDVI , and distance to the nearest freshwater body . The type of freshwater body , ecological zone , and land cover were measured in categorical dimensions . The study year was also included as linear and categorical covariate in order to account for possible temporal trends . The categories were defined on decades as follows: 1980–1989 , 1990–1999 , and from 2000 onwards . For S . mansoni , half of the data were from the 1980s ( 49 . 7% ) , 24 . 1% from the 1990s , whereas 26 . 2% were obtained in the new millennium . For S . haematobium , 37 . 8% of the data stem from the 1980s , 35 . 7% from the 1990s , and 26 . 5% from 2000 onwards . Relevance of continuous or categorized covariates to predict the outcome was assessed based on p-values resulting from likelihood ratio tests ( LRTs ) at significance levels of 0 . 15 . All significant covariates were included in the Bayesian analysis . Bayesian geostatistical logistic regression models were fitted with location-specific random effects . Spatial correlation was modeled assuming that the random effects follow a multivariate normal distribution with variance-covariance matrix related to an exponential correlation function between any pair of locations . Model fit requires the inversion of this matrix . Due to the large number of survey locations in our datasets , parameter estimation becomes unfeasible . An approximation of the spatial process by a subset of survey locations ( ) proposed by Banerjee et al . ( 2008 ) [29] and further developed by Gosoniu et al . ( forthcoming ) [30] and Rumisha et al . ( forthcoming ) [31] was implemented instead . We employed MCMC simulation to estimate the model parameters . Prevalence of infection at 220 , 000 locations was predicted for the most recent decade ( from the year 2000 onwards ) via Bayesian kriging using joint predictive posterior distributions [16] . Due to computational issues , we modeled the multivariate Gaussian spatial process separately for each country . The performance of the models was assessed using model validation via different approaches: mean predictive errors ( ME ) , mean absolute predictive errors ( MAE ) , discriminatory performance on a 50% prevalence cut-off , and Bayesian credible interval ( BCI ) comparisons [17] . Further details pertaining to the Bayesian geostatistical model , sub-sampling , and model validation approaches are given in the Appendix S1 .
A schematic overview of the study profile on obtaining prevalence data on schistosomiasis from the GNTD is given in Figure 1 . The final datasets consisted of 1993 and 1179 survey locations for S . haematobium and S . mansoni , respectively , out of which 1722 and 1094 locations were unique . Observed prevalence of the survey locations ranged from 0% to 100% for each Schistosoma species with mean prevalence of 31 . 0% ( median 15 . 0% , standard deviation ( SD ) 29 . 0% ) for S . haematobium , and 17 . 7% ( median 0 . 0% , SD 24 . 4% ) for S . mansoni . The distribution and the prevalence level of the survey locations are shown in Figures 2 and 3 for S . haematobium and S . mansoni , respectively . An overview of the number of surveys with details given regarding sampling period , diagnostic technique , survey type , and mean prevalence , stratified by country , is given in Table 2 . Spatial distributions of potential covariates influencing the distribution of schistosomiasis are presented in Figure 4 . Bivariate logistic regressions of the continuous factors in relation to the disease outcomes showed that categorical variables predicted better based on BIC values than linear variables for both Schistosoma species ( results not presented ) . Each potential covariate considered for the analyses had a p-value of <0 . 001 based on LRTs and was therefore included in the multivariate analyses . Backwards logistic regressions demonstrated the importance of the whole set of covariates for each species . The resulting odds ratios ( ORs ) of bivariate and multivariate non-spatial logistic regressions are summarized in Table 3 for S . haematobium , and Table 4 for S . mansoni . The only non-significant outcome-predictor relations in a multivariate framework for the former species were yearly averaged precipitation between 300 mm and 399 mm , and NDVI levels between 0 . 33 and 0 . 52 . For the latter species , only altitude levels of at least 500 m above sea level and night LSTs between 20 . 0°C and 20 . 7°C were non-significant . Model parameter estimates for S . haematobium and S . mansoni are presented in Table 3 and Table 4 , respectively . Introduction of spatial correlation led to changes in the significance of covariates and the direction of outcome-predictor relations compared to the corresponding non-spatial multivariate logistic regression models . For example , the influence of rainfall for S . mansoni became more important while the effect of the survey period and non-perennial freshwater bodies was reduced . The spatial range was estimated to be 398 km ( 95% BCI: 384–412 km ) and 387 km ( 95% BCI: 375–402 km ) for S . haematobium and S . mansoni , respectively . These estimates suggest strong spatial correlation for both species . The spatial variation was similar for the two species ( 4 . 02 for S . haematobium vs . 4 . 05 for S . mansoni ) . Figure 5A presents the prevalence map for S . haematobium based on the median of the predictions . Low-prevalence areas ( predicted infection prevalence <10% ) were primarily observed in the Sahara , Cameroon , north-west Côte d'Ivoire , and Senegal . Prevalence >50% are mainly spread along the Niger River , in Sierra Leone , east/central Senegal , and south Nigeria . The map of the SD of model predictions for this species ( Figure 5B ) demonstrates that small prediction errors were primarily found around the survey locations used for sub-sampling . The median spatial S . mansoni prevalence map is shown in Figure 6A with the corresponding error presented in Figure 6B . High-prevalence areas ( predicted prevalence >50% ) were mainly found in north-east Liberia , east Côte d'Ivoire , west Ghana , north/central Benin , west Nigeria , north Cameroon , and central Mali in close proximity to Niger River . Very low prevalence areas ( predicted prevalence <10% ) were predominant in Senegal , The Gambia , Guinea-Bissau , Mauritania , and Niger . Furthermore , low prevalence areas were predicted for north Mali , south Togo , and parts of Cameroon . Areas of high prediction accuracy were found around the sub-sampled survey locations and in desert/semi-desert ecological zones . Table 5 shows population-adjusted country prevalence estimates . For S . haematobium , prevalence estimates range between 17 . 6% ( The Gambia ) and 51 . 6% ( Sierra Leone ) , whereas for S . mansoni they range between 0 . 5% ( The Gambia ) and 37 . 1% ( Liberia ) . S . haematobium was found to be the predominant species throughout West Africa with a difference compared to S . mansoni of up to 30% in Burkina Faso and a minimum difference of about 4% in Liberia . Combined Schistosoma prevalence estimates , assuming independence of the occurrence of the two species , varied from 18 . 1% ( The Gambia ) to 58 . 3% ( Liberia ) with high numbers of infected individuals aged ≤20 years ( more than 5 million ) in Ghana and Nigeria . Lower numbers ( <1 million ) of infected individuals aged ≤20 years were found in The Gambia , Guinea-Bissau , Liberia , and Mauritania . The overall number of infected individuals aged ≤20 years in West Africa is 50 . 8 million . Model validation based on 80% of the survey locations resulted in MEs of −1 . 7 for S . haematobium and 0 . 0 for S . mansoni , and respective MAEs of 19 . 5 and 7 . 3 . The percentage of test locations correctly predicted by 95% BCIs was 72 . 9% for S . haematobium , and 72 . 5% for S . mansoni . ME and MAE comparisons between spatial and exchangeable random effect models showed that spatial models result in better predictive ability ( S . haematobium: ME = 3 . 8 , MAE = 27 . 7; S . mansoni: ME = −0 . 8 , MAE = 14 . 9 ) . Discriminatory performance based on a 50% prevalence cut-off showed that the models correctly predicted 93 . 2% and 76 . 9% of the validation locations for S . mansoni and S . haematobium , respectively . False-high predictions were obtained for 5 . 5% ( S . mansoni ) and 18 . 8% ( S . haematobium ) of the test locations .
To our knowledge , we provide the first model-based prevalence maps for both S . haematobium and S . mansoni for individuals aged ≤20 years in West Africa , including Cameroon . We used a readily available open-access database consisting of a large number of historical and contemporary geolocated and standardized survey data [15] , coupled with Bayesian-based geostatistical tools . Standard geostatistical methods are not able to handle large numbers of survey locations due to computational problems . Therefore , for the first time , an approximation of the spatial process was implemented in Schistosoma prevalence modeling . In comparison to existing prevalence estimates , major shortcomings of previous studies have been addressed , and hence our prevalence maps show a higher spatial resolution and we believe that they are more accurate than heretofore . This claim is justified as follows . First , our estimates are based on the GNTD database that has gone live in July 2010 , developed as part of the EU-funded CONTRAST project . As of February 2010 , the GNTD contained more than 4500 and 2600 unique entries in West Africa for S . haematobium and S . mansoni , respectively . Second , data-tailored statistical methods based on Bayesian geostatistical modeling were used in order to incorporate spatial correlation between survey locations and to obtain more accurate estimates of the uncertainty of the predictions . Third , climatic and environmental covariates were employed in the models to evaluate the effect on the disease outcomes . The climatic and environmental factors were obtained at high spatial resolution to be able to predict small hotspots of risk , which could arise due to the focal distribution of schistosomiasis , which is an important epidemiological feature of the disease [32] . An existing S . haematobium prevalence map for three West African countries ( i . e . , Burkina Faso , Mali , and Niger ) using Bayesian geostatistical modeling was previously presented by Clements et al . ( 2008 ) [33] based on data from 2004–2006 . However , this map does not show the actual level of schistosomiasis prevalence but rather probabilities that the predicted prevalence is above a pre-defined cut-off , arbitrarily set at 50% . This cut-off has been proposed by the World Health Organization ( WHO ) [1] to distinguish between low and high risk areas , and hence such maps are useful to detect areas where preventive chemotherapy might be warranted on an annual basis . However , the maps do not provide detailed information for lower risk areas or the number of infected individuals and they cannot be used for monitoring and evaluation purposes following interventions . A more recent publication by Clements et al . ( 2009 ) [9] presented a S . haematobium prevalence map for the same three West African countries . This map shows similar patterns to our map with the exception of north Burkina Faso . In this area , Clements and colleagues predicted prevalence levels of 10–20% for high and low egg-intensities , while our estimates suggest much higher prevalence ( >50% ) . These discrepancies are most likely due to differences in the underlying survey data . The Clements et al . data were only partially included in the GNTD database as we could not access them fully . The estimated spatial correlation for both Schistosoma species was very strong with spatial ranges of approximately 400 km . Previously reported spatial ranges in parts of West Africa vary between 7 . 5 km [19] and approximately 180 km [33] . However , these estimates were based on recent surveys , and hence influenced by recently established control programs . Interventions are likely to reduce the predictive power of environmental and climatic factors on the distribution of schistosomiasis and , thus , reduce spatial correlation . Similar effects were found for malaria , where historic data showed stronger spatial correlation [34] than recent surveys [35] , [36] . We overlaid population data adjusted to 2010 on the predicted prevalence surfaces for the two Schistosoma species in order to obtain country-specific estimates of the number of infected individuals aged ≤20 years . Previous country estimates , for instance those presented by Chitsulo et al . ( 2000 ) [4] , Steinmann et al . ( 2006 ) [2] , or Utzinger et al . ( 2009 ) [5] , are interpolations of limited observations for a whole country , and hence lack empirical modeling . Chitsulo and colleagues reported a higher number of infected people for West Africa ( 71 . 8 million ) compared to our estimate ( 50 . 8 million ) . Of note , the Chitsulo et al . estimates are based on the whole population , while our new estimates concern the age group ≤20 years . Moreover , the Chitsulo et al . estimates pertain to mid-1990s population estimates , compared to our adjusted estimates for the year 2010 . In countries like Cameroon , The Gambia , Ghana , and Liberia , characterized by high rural-to-urban migration in the last decade , the Chitsulo et al . prevalence estimates should be treated with care due to rapid urbanization . Our study revealed that the combined prevalence of S . haematobium and S . mansoni in The Gambia , for example , is two-fold lower than previously reported by Chitsulo et al . ( 18 . 1% vs . 37 . 5% ) . However , in Benin , Guinea , Liberia , Nigeria , and Togo , we found prevalence estimates that are more than 10 percentage points higher than the previous estimates . On the one hand , differences might be related to sparse data , for example , in Benin , The Gambia , Guinea , Guinea-Bissau , Liberia , Mauritania , Nigeria , and Sierra Leone . Previous estimates failed to take into account model-based predictions on the basis of climate , environment and disease data . Since we modeled disease prevalence on individuals aged ≤20 years ( highest risk groups ) , the prevalence estimates correspond to the former risk group . Therefore they are likely to overestimate the prevalence in the whole population . We estimated the country-specific overall schistosomiasis prevalence by assuming independence between the occurrence of S . haematobium and S . mansoni in each area . However , it is conceivable that simultaneous infections with both species is more frequent than expected by chance in areas where the species co-exist as infection pathways are similar and highly behavioral related . Hence , the combined prevalence estimates potentially underestimate the true schistosomiasis situation in West Africa . A modeling approach via joint spatial random effects [37] could assess the effect of potential dependence between the species , but would increase the number of spatial parameters and is therefore computationally challenging . We might also underestimate schistosomiasis prevalence in Cameroon , Mali , and Nigeria because of the presence of S . intercalatum [4] . We did not include this species in the analysis since the GNTD database currently only contains 17 survey locations outside Cameroon . However , it is assumed that S . intercalatum has a low prevalence [4] and there are signs that this species is further declining in importance [38] . Model validation has shown that the S . haematobium predictions seem to overestimate the actual prevalence , while the S . mansoni model revealed no tendency to over- or underestimate the overall prevalence . The MAE for the S . haematobium model is nearly three times larger than the one for S . mansoni . This is expected because the mean prevalence for S . haematobium was about double than that for S . mansoni . Our models correctly predict about 72% of the survey locations when considering 95% BCIs . We are encouraged by these results , since perfect predictions are rather unlikely in reality due to the complexity of disease transmission . However , our models are based on assumptions , which could influence model performance . We assumed that the diagnostic techniques employed have similar ability to detect an infection , but different diagnostic techniques show differences in sensitivity and specificity , which also depends on the overall prevalence and infection intensity [39] . This might have led to an underestimation of prevalence due to the imperfect sensitivity of direct diagnostic techniques [39] . Additional model parameters accounting for the performance of the different diagnostic techniques could be incorporated in the models . However in the absence of detailed information regarding sampling effort , assumptions would be required which may be debatable and introduce additional biases . We are currently examining the effect of different approaches on addressing this issue on the model-based predictions . We did not adjust the outcome according to age and sex even though the age groups differ and especially school surveys are likely to include more boys than girls due to prevailing cultural issues in many parts of West Africa . Therefore , our results are likely to be biased and potentially overestimate schistosome prevalence . However , many publications do not present stratified results by these subgroups . Age-adjustment models are feasible but difficult to implement because age-prevalence curves have to be fitted for different transmission settings [40] . Furthermore , disease data are often reported at wide age ranges ( i . e . , school-aged children ) and individuals might not be well distributed within the range introducing bias even though an age-prevalence model is taken into account . Surveys are typically conducted in endemic areas leading to high observed prevalence levels . This could result in an overestimation of prevalence in the present analysis . However , in the data we analyzed , 45% of the locations for S . haematobium and 73% for S . mansoni had an observed prevalence levels below 10% . We therefore assume that a location selection bias is unlikely . Another concern is the large amount of zero outcomes ( i . e . , none of the study participants found to be infected ) especially for S . mansoni ( S . mansoni: 54 . 1%; S . haematobium: 20 . 1% ) . To overcome this issue , zero-inflated models need to be incorporated , which modify the likelihood function and add an additional model parameter capturing the over-dispersion arising by the zeros [41] . The models presented in this manuscript did only include spatial random errors , and hence we ignored potential measurement errors . Inclusion of location-specific non-spatial error terms might have improved model predictions . However , location-specific non-spatial error terms would have doubled the number of error terms leading to highly parameterized models . We further assumed isotropic stationary models . Non-stationary models imply that the spatial random effect is varying from one region to another and is not stable throughout the study area [35] . This assumption has been confirmed by semi-variogram comparisons showing that the estimated spatial range parameters for S . mansoni differ between eco-zones . However , semi-variogram analyses did not indicate non-stationarity in the spatial distribution of S . haematobium . Isotropic models assume that the spatial correlation is the same within the same distance irrespective of direction [42] . This assumption might not be valid since intermediate host snails spread along rivers and lakeshores and , therefore , introduce correlation attributed to directions . The choice and size of sub-sampled locations required to adequately approximate the spatial Gaussian process is a research area on its own in spatial statistics . Many different approaches are available to optimize selection . We implemented a method based on semi-variogram comparisons . This selection is aiming to preserve the spatial surface of the original dataset . However , it might fail to identify a sub-sample , which minimizes the prediction error . The spatially averaged predictive variance ( SAPV ) method proposed by Finley is trying to optimize the variance in the predictions , but implementation is computationally highly demanding [43] . Time-dependent covariates , such as the climatic factors , might have changed between the 1980s and the 2000s . However , our geographical covariates were solely based on recent remote sensing data ( from 2000 onwards ) , because historical remote sensing data are , to our knowledge , not freely available at high spatial and temporal resolution . The long run averages of the recent data enable us to maintain high spatial resolution although they cannot capture variation in the observed outcome due to unusual climatic conditions or climate change that might have occurred since the 1980s and 1990s . Preliminary residual analyses suggest that there is only weak temporal correlation in the data . We therefore only modeled a spatial rather than a spatio-temporal process . This led to a more parsimonious model and facilitated model fit . Nevertheless , we incorporated temporal trends in the prevalence estimation by including the survey year as covariate . Both Schistosoma species showed that the predicted prevalence was highest during the 1990s . This increase might be explained by water resources development and management activities ( e . g . , the construction of dams and irrigation systems ) , political unrests and civil restructuring . Water resources development and management projects might have improved the suitability of the environment for snail intermediate hosts that might have spread into previously snail-free zones together with the parasites . Since the beginning of the new millennium , a number of large-scale preventive chemotherapy programs are underway in parts of West Africa and it will be important to monitor how the prevalence of schistosomiasis changes in space and over time . The effectiveness of control interventions may vary across areas but , to our knowledge , a comprehensive database compiling this information with high spatio-temporal resolution has yet to be established . Concluding , our country-specific Schistosoma prevalence estimates and numbers of individuals aged ≤20 years infected with either S . mansoni , or S . haematobium , or both species concurrently presented here are useful tools for disease control managers and other stakeholders to support decision-making on interventions . Our maps can also serve as a benchmark to monitor the impact of control interventions and for long-term evaluation on transmission dynamics . Model-based estimates in areas with scarce data and high uncertainty could be improved by additional surveys to enhance our knowledge on the distribution of schistosomiasis and disease burden . We plan to further expand this work to other regions and address the issues of non-stationarity , diagnostic sensitivity , and age-heterogeneity across surveys . Finally , we will test the assumption of independence between the Schistosoma species to improve accuracy of the joint prevalence estimates . | Schistosomiasis is a parasitic disease caused by a blood fluke that mainly occurs in Africa . Current prevalence estimates of schistosomiasis are based on historical data , and hence might be outdated due to control programs , improved sanitation , and water resources development and management ( e . g . , construction of large dams and irrigation systems ) . To help planning , coordination , and evaluation of control activities , reliable schistosomiasis prevalence estimates are needed . We analyzed compiled survey data from 1980 onwards for West Africa , including Cameroon , focusing on individuals aged ≤20 years . Bayesian geostatistical models were implemented based on environmental and climatic predictors to take into account potential spatial clustering within the data . We created the first smooth data-driven prevalence maps for Schistosoma mansoni and S . haematobium at high spatial resolution throughout West Africa . We found that an estimated 50 . 8 million West Africans aged ≤20 years are infected with schistosome blood flukes . Country prevalence estimates ranged between 0 . 5% ( in The Gambia ) and 37 . 1% ( in Liberia ) for S . mansoni and between 17 . 6% ( in The Gambia ) and 51 . 6% ( in Sierra Leone ) for S . haematobium . Our results allow prioritization of areas where interventions are needed , and to monitor and evaluate the impact of control activities . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [
"medicine",
"disease",
"mapping",
"public",
"health",
"and",
"epidemiology",
"epidemiology",
"spatial",
"epidemiology"
] | 2011 | Geostatistical Model-Based Estimates of Schistosomiasis Prevalence among Individuals Aged ≤20 Years in West Africa |
Aggregation is a social behavior that varies between and within species , providing a model to study the genetic basis of behavioral diversity . In the nematode Caenorhabditis elegans , aggregation is regulated by environmental context and by two neuromodulatory pathways , one dependent on the neuropeptide receptor NPR-1 and one dependent on the TGF-β family protein DAF-7 . To gain further insight into the genetic regulation of aggregation , we characterize natural variation underlying behavioral differences between two wild-type C . elegans strains , N2 and CB4856 . Using quantitative genetic techniques , including a survey of chromosome substitution strains and QTL analysis of recombinant inbred lines , we identify three new QTLs affecting aggregation in addition to the two known N2 mutations in npr-1 and glb-5 . Fine-mapping with near-isogenic lines localized one QTL , accounting for 5%–8% of the behavioral variance between N2 and CB4856 , 3′ to the transcript of the GABA neurotransmitter receptor gene exp-1 . Quantitative complementation tests demonstrated that this QTL affects exp-1 , identifying exp-1 and GABA signaling as new regulators of aggregation . exp-1 interacts genetically with the daf-7 TGF-β pathway , which integrates food availability and population density , and exp-1 mutations affect the level of daf-7 expression . Our results add to growing evidence that genetic variation affecting neurotransmitter receptor genes is a source of natural behavioral variation .
Most animal and human behaviors are variable , in part due to genetic variation between individuals . Genetic mapping of strain differences in anxiety , learning , activity levels , and the response to addictive drugs in mice , as well as aggression , locomotor activity , and sensory behaviors in Drosophila , has demonstrated a complex genetic basis for these traits [1]–[3] . Typically , multiple loci contribute to each behavior , the contribution of each locus is small , and the effect of many individual loci depends on the genotype at other loci and on environmental conditions [1] . This genetic complexity poses challenges for the discovery of specific genetic variants that modulate behavior , and consequently only a few genes contributing to natural behavioral diversity have been definitively identified . Defining quantitative behavioral genes at a molecular level has the potential to point to classes of genes that generate behavioral variation and to provide new insights into the neuronal control of behavior . Social behaviors are central to the survival and reproductive success of humans and animals , and defects in social cognition and interaction are core features of human autism and schizophrenia [4] , [5] . Social behaviors are also variable within and between animal species , providing a starting point for genetic analysis . Animals within a species display different social behaviors based on their sex , developmental stage , reproductive status and environmental conditions [6] . In addition , these behaviors are shaped by individual genetic variation that interacts with environmental factors . For example , genetic variation among Drosophila males affects their territorial aggressive behavior , and this genetic variation interacts with an environmental regulator of aggression , population density [7]–[9] . Aggregation between members of a species is a social interaction that can provide direct benefits , such as the conservation of body heat [10] , as well as facilitating more complex behaviors such as reproduction , migration , defense , or communal foraging [11] . However , aggregation increases competition for local resources and facilitates disease transmission , so it is not always favorable [11] , [12] . Accordingly , many animals aggregate under certain conditions but not others . Aggregation behavior and its regulation by environmental and genetic factors have been studied extensively in the nematode Caenorhabditis elegans . C . elegans aggregates spontaneously on food at high population density [13] , a behavior that is enhanced at high oxygen levels [14] , when food is depleted [15] , or under other stressful conditions . Similar behaviors are observed in other nematode species at high density [16] . The role of aggregation in C . elegans biology is poorly understood , but it might promote mating [17] , lower oxygen to a preferred intermediate level [14] , or expose young animals to pheromones that drive entry into the stress-resistant dauer larva stage [17] . Different wild-type C . elegans strains vary in their propensity to aggregate . The “solitary” laboratory strain N2 aggregates infrequently on a lawn of bacterial food , whereas wild-caught “social” strains aggregate at much higher rates [13] , [18] . In addition to aggregating , wild social animals move quickly on food , have a stronger preference for certain pheromones , and accumulate on the oxygen-poor border of a bacterial lawn; the latter behavior is called bordering [14] , [18] , [19] . Aggregation and bordering typically occur under the same conditions , in part because oxygen regulates both behaviors [13]–[15] , [18]–[23] . The extreme solitary behavior of the N2 strain arose as an adaptation to laboratory conditions [24] , and results from a gain-of-function point mutation in the neuropeptide receptor gene npr-1 [18] together with a loss-of-function rearrangement of the sensory globin gene glb-5 [24] , [25] . Both NPR-1 and GLB-5 act in a circuit that regulates sensitivity to aggregation-promoting environmental signals . GLB-5 acts in sensory neurons to regulate sensitivity to environmental oxygen [24] , [25] , whereas NPR-1 acts in an interneuron hub of a circuit that integrates aggregation-promoting sensory cues from oxygen , noxious chemicals , and pheromones [19] . Thus the genetic and environmental regulation of aggregation converge on a shared neuronal circuit to regulate behavior . A second genetic pathway that regulates aggregation is controlled by the TGF-β homolog DAF-7 , which serves a key neuroendocrine role in integrating nutrient availability , overpopulation stress , and physiology in C . elegans [20] . The presence of food induces daf-7 expression in the ASI sensory neurons , whereas population density pheromones suppress daf-7 in ASI [26] , [27] . DAF-7 protein is secreted from ASI and activates the TGF-β receptors DAF-1 and DAF-4 in a variety of neurons to regulate dauer larva development , aggregation , fat accumulation , gene expression , and lifespan [20] , [21] , [26]–[30] . Loss-of-function mutations in daf-7 or its receptors result in aggregation and bordering in the N2 genetic background , and aggregation is enhanced in daf-7; npr-1 double mutants , suggesting that these two pathways act at least partly independently of one another [20] , [21] . By characterizing intercrosses between the N2 laboratory strain and the Hawaiian strain CB4856 , we observed variability in aggregation and bordering behavior that were not explained by the laboratory-induced npr-1 and glb-5 mutations . This variability suggested the existence of additional quantitative trait loci for these behaviors . Here we use a set of recombinant inbred advanced intercross lines ( RIAILs ) and chromosome substitution strains to probe the genetic architecture and molecular basis of natural variation in aggregation and bordering behavior between the two C . elegans strains , structuring the analysis to control for the strong effect of npr-1 . We show that aggregation and bordering are genetically complex , with multiple contributing quantitative trait loci ( QTLs ) , and refine one QTL to identify a new gene affecting the daf-7 pathway , the GABA receptor EXP-1 .
The Hawaiian CB4856 ( HW ) C . elegans strain is highly divergent from the N2 laboratory strain at many loci [31] , including the neuropeptide receptor gene npr-1 . To probe the combined effects of loci other than npr-1 on aggregation and bordering behavior , we examined two near-isogenic lines ( NILs ) that differed from each parental strain by the genetic substitution of a small region near the npr-1 gene ( Figure 1A , 1B; see Methods ) . As shown previously , a NIL in which the HW allele of npr-1 was introduced into an N2 background resulted in increased levels of aggregation and bordering [18] , [24] ( Figure 1B ) . These levels were , however , significantly lower than those of the HW strain under the conditions examined . Conversely , introducing the N2 allele of npr-1 into the HW background did not restore behavior to N2-like levels , instead resulting in a strain with intermediate levels of aggregation and bordering ( Figure 1A , 1B ) . These results indicate that loci in addition to npr-1 affect aggregation and bordering behaviors . To systematically define genetic differences between N2 and HW , we characterized chromosome substitution strains ( CSS ) in which each of the six N2 chromosomes was individually replaced by a HW chromosome [32] . Strains bearing HW chromosomes II or V had significantly higher bordering than N2 , and the strain bearing the HW X chromosome had high levels of both bordering and aggregation , identifying three chromosomes with loci affecting these behaviors ( Figure 1B ) . The known laboratory-derived mutations in npr-1 and glb-5 are on X and V , respectively . However , the effects of X chromosome substitution were significantly greater than those of the HW npr-1 NIL , and effects of chromosome V substitution were significantly greater than those of a similar NIL bearing the HW allele of glb-5 . These results imply the existence of at least one additional QTL on each of chromosomes V and X ( Figure 1B ) . Thus the combination of CSS and NIL analysis indicates that aggregation and bordering are affected by at least five loci that differ between N2 and HW: one or more loci on II , glb-5 and at least one additional locus on V , and npr-1 and at least one additional locus on X . In a parallel approach , QTL analysis was performed on recombinant inbred advanced intercross lines ( RIAILs ) derived from crosses between N2 and HW [31] . To set aside the large effect of the npr-1 mutation , we examined only strains with the N2 allele of npr-1 . These 102 RIAILs had a continuous quantitative distribution of bordering and aggregation behaviors , implying the existence of multiple QTLs rather than one locus of large effect ( Figure 1C ) . Bordering and aggregation behaviors were strongly but not perfectly correlated in the RIAILs ( r = 0 . 73 , 99%C . I . = 0 . 58–0 . 83 ) , suggesting that genetic contributions to bordering and aggregation in these strains are similar but perhaps not identical . QTL analysis of the RIAILs identified a significant QTL on chromosome II ( II-QTL ) for aggregation and a significant QTL on chromosome V ( V-QTL ) for bordering ( Figure 1D ) . The II-QTL explains 8 . 2% ( P<0 . 01 ) of the aggregation variance and 5 . 3% ( P = 0 . 019 ) of the bordering variance in the RIAILs , whereas the V-QTL explains 14% ( P<0 . 01 ) of the bordering variance and an insignificant fraction of the aggregation variance ( see Methods ) . The chromosome V QTL overlaps glb-5 , as well as covering a broader region that may encompass the second bordering QTL inferred from the chromosome V substitution strain ( Figure 1B , 1D ) . The II-QTL does not correspond to a previously characterized locus , and was analyzed further . The QTL analysis of RIAILs placed the II-QTL near 6 . 62 Mb ( 1 . 5-LOD support interval = 2 . 48–9 . 55 Mb ) . To confirm this map position , we created a NIL , kyIR20 , containing the peak of the HW II-QTL ( from 4 . 77 to 6 . 65 Mb ) in an N2 background . The behavior of kyIR20 resembled that of the chromosome substitution strain bearing all of HW chromosome II , with 2 . 5-fold more bordering than the N2 strain , and a small effect on aggregation ( Figure 2A ) . The effects of the II-QTL on bordering and aggregation cosegregated throughout fine-mapping of the QTL , suggesting that they have a common genetic basis ( see below ) . To identify the specific HW region ( s ) that affect behavior , we generated recombinants between N2 and the kyIR20 NIL , deriving NILs with less HW DNA than kyIR20 . These NILs defined an interval from 6 . 047 to 6 . 192 Mb , a region encompassing 56 genes , as a minimal region sufficient to promote bordering and aggregation ( Figure 2A ) . A NIL that behaved significantly different from N2 , kyIR97 , contained only 270 kb of HW DNA ( Figure 2A ) . To fine-map this II-QTL , 5000 F2 progeny of crosses between the kyIR97 NIL and N2 were screened at the DNA level for recombination events within the QTL interval , and the recombinants were tested for aggregation and bordering behaviors . Five informative recombination events split this region . The behavior of the five recombinants indicated the presence of genetic changes necessary for bordering and aggregation in a 6 . 2 kb interval ( 6 , 148 , 781–6 , 154 , 990 ) ( Figure 2B , 2D ) . kyIR110 , a NIL with less than 45 kb of HW DNA covering the 6 . 2 kb interval , retained higher bordering and aggregation behaviors than N2 ( Figure 2C ) . These genetic results definitively placed a QTL within the 45 kb associated with kyIR110 , and support a location for one quantitative trait nucleotide within the smaller 6 . 2 kb interval . However , a trend toward smaller behavioral effects as the QTL was refined leaves open the possibility that additional causative polymorphisms between N2 and HW are present in the larger kyIR20 interval . The 6 . 2 kb minimal QTL interval fell within a single gene , abts-3 , which encodes a predicted anion transporter ( Figure 2D ) . Sequencing this region uncovered 11 polymorphisms between HW and N2 ( Figure 2D and Table S1 ) : five noncoding single nucleotide polymorphisms ( SNPs ) , two coding SNPs ( abts-3a G615D , abts-3d Q118P ) , one single nucleotide deletion , one single nucleotide insertion , a three-nucleotide single amino acid insertion ( abts-3d 115I116 ) , and a 23-nucleotide deletion in HW . Sequence analysis of 59 additional wild strains indicated that 10 of the 11 sequence polymorphisms were represented in other wild C . elegans populations , whereas one SNP was present only in HW ( Table S1 and data not shown ) . Notably , the N2 sister strain LSJ2 , which separated from N2 soon after their isolation from the wild [24] , was identical in sequence to N2 at this locus . Thus the N2 sequences are likely to be present in wild populations , and not laboratory-derived . The existence of one private polymorphism in HW is consistent with the fact that HW is one of the most divergent wild C . elegans strains , harboring many polymorphisms not found in any other strain [33] . The mapping of the II-QTL defined the location of the relevant sequence change , but not necessarily the affected gene , as noncoding regulatory changes could act at a distance to affect neighboring genes [34]–[38] . To define the gene affected by the II-QTL , we performed quantitative complementation tests between the II-QTL and loss-of-function mutations in genes in the region [1] , [39] . In this test , the N2 and HW QTLs were examined as heterozygotes with null alleles of candidate genes , with the expectation that the null allele would fail to complement the QTL with reduced activity . Initial experiments indicated that the bordering and aggregation behaviors of an N2 II-QTL/HW II-QTL heterozygote resembled the N2 II-QTL homozygote ( Figure 3A ) . The recessive nature of the HW II-QTL suggested that the HW phenotype should be observed in a HW/null heterozygote . Since the 6 . 2 kb minimal QTL interval was fully contained within abts-3 , this gene was considered the most promising candidate . However , a deletion allele of abts-3 ( ok368 ) complemented the HW II-QTL , as well as the N2 II-QTL , to give N2-like behavior in heterozygotes ( Figure 3B; the location of the mutation is shown in Figure 2D ) . This result argued against abts-3 being the gene affected by the II-QTL . A second gene close to the II-QTL is exp-1 , which encodes a γ-aminobutyric acid ( GABA ) -gated cation channel [40]; the stop codon of exp-1 is 2 . 2 kb away from the 6 . 2 kb QTL ( Figure 2D ) . In a quantitative complementation test , a loss-of-function mutation in exp-1 ( ox276 ) failed to complement the HW II-QTL , with the heterozygote showing substantial bordering and aggregation behaviors ( Figure 3B ) . Control experiments demonstrated that the exp-1 mutation was fully complemented by the N2 II-QTL , excluding dominant effects of exp-1 ( Figure 3B ) . Two-way ANOVA provided strong statistical support for an interaction between the II-QTL and exp-1 , but not abts-3 ( Figure 3C ) . These results suggest that exp-1 is a quantitative trait gene that affects bordering and aggregation . To explain these results , we suggest that noncoding variation 3′ of the exp-1 transcript , within the abts-3 gene , modifies aggregation and bordering behavior ( at least in part ) by affecting the activity of exp-1 . The overall abundance of exp-1 mRNA measured by quantitative RT-PCR was similar in N2 and in the HW II-QTL strain ( Figure S1 ) , suggesting that the 3′ sequences in the II-QTL may confer specific spatial or temporal patterns of expression , rather than affecting total mRNA levels . As an independent test of the role of exp-1 in social behavior , we examined five exp-1 mutant alleles in an N2 genetic background: a predicted null allele , exp-1 ( ox276 ) , and four missense alleles , exp-1 ( n2570 ) , exp-1 ( n2641 ) , exp-1 ( n2676 ) , and exp-1 ( sa6 ) [40] . All exp-1 mutants had high levels of bordering behavior , and exp-1 ( ox276 ) , exp-1 ( n2570 ) , and exp-1 ( n2641 ) had significantly increased aggregation ( Figure 3D ) . In a complementation test , exp-1 ( ox276 ) /exp-1 ( sa6 ) trans-heterozygotes failed to complement for bordering behavior , as expected if they affect the same complementation group ( gene ) ( Figure 3D ) . These results indicate that normal exp-1 activity suppresses bordering and aggregation in the N2 strain , and suggest that the II-QTL from HW has a reduced level of exp-1 activity . Previous studies of exp-1 in the enteric nervous system defined a region sufficient for rescue of exp-1 phenotypes related to defecation [40] , but this clone did not include the 6 . 2 kb downstream region defined by the II-QTL . A transgene containing the minimal exp-1 region involved in defecation failed to rescue the bordering and aggregation defects of the exp-1 ( ox276 ) deletion mutant ( Figure S2 , transgene ‘6’ ) , suggesting that it lacked regulatory sequences for exp-1 expression relevant to the bordering phenotype . Among multiple tested transgenes spanning the exp-1 locus and adjacent sequences , only transgenes covering 70 kb encompassing both exp-1 and abts-3 effectively rescued the social behaviors of exp-1 ( ox276 ) ( Figure S2 , transgene ‘4+5’ ) . Similar transgenes reduced bordering in the HW II-QTL strain ( Figure S2 ) . These results suggest that exp-1 regulation of aggregation and bordering requires long-range regulation , including substantial sequences 3′ of exp-1 ( see Discussion ) . exp-1 encodes an unconventional GABA-gated cation channel that depolarizes cells in the presence of GABA [40]; it is one of four genetically characterized C . elegans GABA receptors . All GABAergic signaling in C . elegans is blocked by mutations in unc-25 [41] , [42] , which encodes the GABA biosynthetic enzyme glutamate decarboxylase , so unc-25 mutants would be expected to have phenotypes related to those of exp-1 . Indeed , unc-25 ( n2324 ) mutants showed bordering behaviors , albeit weaker than those of the exp-1 deletion mutant ( Figure 4A ) . An exp-1 ( ox276 ) ; unc-25 ( n2324 ) double mutant did not have enhanced defects , and in fact resembled the milder unc-25 mutant rather than the stronger exp-1 mutant ( Figure 4A ) . The shared effects of exp-1 and unc-25 are consistent with the hypothesis that GABA regulates bordering behavior by activating exp-1 . The milder phenotype of unc-25 and the double mutants suggests that GABA might also activate a second GABA receptor with an opposite effect to exp-1 . The daf-7 neuroendocrine pathway , which is regulated by food and pheromone levels , inhibits aggregation in the N2 genetic background [20] . In the standardized aggregation assay used here , daf-7 ( e1372 ) mutant adults showed high levels of aggregation and low levels of bordering compared to exp-1 mutants ( Figure 4B ) . Adult exp-1 ( ox276 ) ; daf-7 ( e1372 ) double mutant animals did not have enhanced behavioral phenotypes compared to single mutants , suggesting that the two genes act in a common pathway ( Figure 4B ) . In agreement with this possibility , a mutation in the downstream daf-3 co-SMAD transcriptional regulator , which suppresses daf-7 aggregation phenotypes [20] , also suppressed the bordering and aggregation of exp-1 ( Figure 4B ) . The relatively low levels of bordering in daf-7 adults were unexpected based on the literature [20] , but may be explained by developmentally-regulated differences in behaviors: we observed that L4-stage daf-7 animals bordered and aggregated more extensively than adult animals ( Figure S3 ) , and previous studies did not analyze these stages separately . In L4 animals , as in adults , the bordering and aggregation of exp-1 ( ox276 ) ; daf-7 ( e1372 ) double mutant animals was no stronger than that of the most severe single mutant , suggesting participation in a common pathway ( Figure S3 ) . Transcription of daf-7 is regulated by environmental conditions: daf-7 mRNA levels increase when food is abundant and decrease in the presence of density pheromones [26] , [27] . This transcriptional regulation is partly intrinsic to the ASI sensory neuron that expresses daf-7 , and partly determined by intercellular signaling between other sensory neurons and ASI [43] . Using RT-PCR , we asked whether daf-7 transcriptional regulation was altered in exp-1 mutants . exp-1 ( ox276 ) mutant animals had a 4-fold decrease in daf-7 mRNA levels compared to controls ( Figure 4C ) . Moreover , daf-7 mRNA levels in the minimal II-QTL strain kyIR110 were intermediate between those of N2 and those of exp-1 mutants , consistent with the possibility that the II-QTL causes a partial reduction of exp-1 activity ( Figure 4C ) . In combination with the genetic epistasis results , these observations suggest that exp-1 and the II-QTL stimulate aggregation and bordering at least partly through effects on daf-7 expression ( Figure 4D ) .
Using chromosome substitution strains and RIAILs between N2 and HW C . elegans strains , we found that at least five QTLs affect bordering and aggregation behavior . Two QTLs correspond to the known laboratory-acquired mutations in npr-1 and glb-5 , two remain to be identified , and one QTL maps to a region near the abts-3 and exp-1 genes . The minimal 6 . 2 kb QTL is entirely contained within abts-3 , but our results argue that the transcription unit affected by the QTL is the neighboring gene exp-1 , which encodes a GABA receptor . exp-1 and abts-3 genes are adjacent and convergently transcribed in all five Caenorhabditis species for which data is available ( www . wormbase . org ) , suggesting that this synteny might be functionally relevant . Both association studies and linkage studies map causal genetic variants ( i . e . functionally relevant sequence polymorphisms ) , but if the causal variants affect regulatory sequences , they can be located far from the affected gene or genes [1] , [38] . The exp-1 transcript ends 2 . 2 kb from the nearest polymorphism in the II-QTL , suggesting that the QTL affects a regulatory region 3′ to the exp-1 coding region . Most known C . elegans regulatory sites are upstream of or within coding regions , but there are precedents for 3′ transcriptional regulatory elements , including an element located 5 . 6 kb 3′ of the egl-1 cell death gene [37] and elements located 3′ to the osm-9 and ocr-2 family of sensory ion channel genes [44] , [45] , and there are also many mammalian precedents for 3′ regulation [46] , [47] . In the context of natural variation , a recent study identified a 3′ regulatory region of the Nasonia unpaired-like gene that affects wing width differences between wasp species [48] . 3′ regulatory regions were not represented in over 300 other genetic variants that cause phenotypic variation within and between species; this may represent either a real difference or an ascertainment bias toward 5′ regulatory elements ( studies compiled in [49] , [50] ) . Human SNPs that affect gene expression ( eQTNs ) are three times more abundant 5′ of transcriptional start sites than 3′ of transcriptional end sites , and human enhancers are also more commonly 5′ of the gene they regulate , suggesting that the bias toward 5′ regulatory elements is real but not absolute [51] , [52] . Most C . elegans mutations can be rescued by transgenes that cover relatively short regions surrounding the gene of interest . By contrast , the aggregation and bordering behaviors of the II-QTL and the exp-1 ( ox276 ) mutant were refractory to rescue with small transgenes that could rescue the enteric nervous system defects of the exp-1 mutation [40] . The only genomic DNA fragments that successfully rescued exp-1 ( ox276 ) aggregation and bordering covered 26 kb 5′ and 40 kb 3′ of the exp-1 coding region , including all abts-3 transcripts and several additional transcripts . One possible explanation for this result is that appropriate exp-1 expression requires several long-range cis-regulatory elements , including the 6 . 2 kb QTL and other distal sequences . A second possibility is that exp-1 is unusually sensitive to general DNA context , resembling genes expressed in the C . elegans germline , which are transcriptionally silenced unless they are embedded in a complex genomic context [53] . exp-1 encodes an unconventional GABA-gated cation channel that depolarizes cells rather than inhibiting them [40] . It was first identified based on its excitatory action on enteric muscles , but exp-1–GFP fusions are also expressed in some neurons ( PDA , RID , ADE , SABD ) [40] . The relevant site of exp-1 expression for social behavior is unknown , because the minimal transgene that rescues the enteric defect did not rescue social behavior , and therefore may not encompass the full expression pattern of exp-1 . A fuller characterization of the 70 kb genomic region that does rescue social behavior may provide insights into sites of exp-1 expression . GABA-deficient animals have mild bordering and aggregation behavior phenotypes , consistent with a likely function of exp-1 as a GABA receptor . GABA is produced by just 26 neurons in C . elegans; identification of the relevant GABAergic neurons may assist in defining the exp-1 circuit . exp-1 and the daf-7 TGF-β pathway show genetic interactions , suggesting a common role in aggregation . In agreement with this interaction , the level of daf-7 mRNA was reduced in exp-1 null mutants and in animals with the HW exp-1 QTL . exp-1 may be part of the system for detecting environmental stresses that regulates daf-7 expression and neuroendocrine function . The daf-7 pathway and the GABA neurotransmitter system also cooperate to regulate C . elegans dauer development [54] , perhaps by using the same transcriptional mechanism defined here . These results add to evidence that genetic polymorphisms affecting neurotransmitter receptors are sources of natural behavioral variation , and reinforce the importance of studying natural variation as a means toward new biological insights . In humans , very few genetic variants that affect behavioral traits have been mapped . Among these are the ligand-gated ion channels CHRNA3 and CHRNA5 , which encode nicotinic acetylcholine receptor subunits implicated in cigarette smoking behavior by genome-wide association studies [55] . Thus ligand-gated ion channels represent sites of behavioral variation both in C . elegans and in humans . Previous studies in C . elegans and in humans have suggested that G protein-coupled neurotransmitter receptors may be preferred sites of behavioral variation . Among the G protein-coupled receptors implicated in behavioral variation are the C . elegans tyramine receptor tyra-3 , which modifies exploratory behavior [56] , the C . elegans neuropeptide receptor npr-1 , which affects social behavior [18] , and human variation in receptors for serotonin , dopamine , and several neuropeptides that has been associated with psychiatric traits [57]–[60] . Both G protein-coupled receptors and nicotinic acetylcholine receptors in the brain are considered modulatory because they are not essential for fast neurotransmission [61] , and in both cases the receptors belong to large families of related genes that could provide a reservoir for genetic variation . These properties may allow polymorphisms in modulatory receptors to generate variation in neuronal circuits while sparing the core functions of the nervous system . Our results demonstrate the ability to identify such genetic variants even when they have small quantitative effects on behavior .
Strains were grown and maintained under standard conditions at 22–23°C ( room temperature ) on Nematode Growth Medium ( NGM ) 2% agar plates [62] . All animals used for behavioral assays were grown on plates seeded with Escherichia coli OP50 . Aggregation and bordering behaviors were measured essentially as described [18] , with modifications from [19] . Briefly , 2–3 week old 2% agar NGM plates ( stored at 4°C ) were seeded with 200 µL of a saturated E . coli OP50 bacterial culture in LB 2 days before the assay and left at room temperature . 150 adult animals were picked onto the assay lawn . After two hours at 22–23°C , bordering and aggregation behavior were quantified by eye using a dissecting microscope . An animal was considered to be bordering if its whole body resided within 1 mm of the border of the bacterial lawn . Aggregation behavior was measured as the fraction of animals that were in contact with two or more other animals along at least 50% of their body length; this criterion is highly stringent but unambiguous . Each strain was tested at least five times , except for CB4856 ( Figure 1B , four assays ) , RIAILs ( Figure 1C , at least three assays each ) , and transgenic rescued lines ( Figure S3 , at least three assays per line , and at least three lines per tested clone ) . Introgression lines ( Figure 2 ) were tested at least seven times . The N2-HW recombinant inbred advanced intercross lines ( RIAILs ) used in this study represent the terminal generation of a 20-generation pedigree founded by reciprocal crosses between N2 and HW . The lines were constructed through 10 generations of intercrossing followed by 10 generations of selfing [31] . They have been genotyped at 1454 nuclear and one mitochondrial markers and have a 5 . 3-fold expansion of the F2 genetic map [31] . Each RIAIL was tested at least three times . QTL analysis was performed on the mean bordering and aggregation of N2-HW RIALs by nonparametric interval mapping at 1 cM intervals in R/qtl [63] . Significance levels were estimated from 10 , 000 permutations of the data . Percent variance explained by QTLs was based on QTLs defined by the marker with the highest LOD score in the II-QTL ( which reached genome-wide significance for aggregation ) and the V-QTL ( which reached genome-wide significance for bordering ) . Percent variance explained was calculated for each QTL with ANOVA . While it is not common to measure effect sizes for traits for which a statistically significant QTL is not found , we report the effect sizes of the II-QTL and the V-QTL for both bordering and aggregation behaviors in the text because the two traits have a strong genetic correlation . Near-isogenic lines were created by backcrossing a chromosomal region or allele into the desired genetic background as described below . Desired segments were then inbred to homozygosity . Marker positions are based on Wormbase release WS229 . CX11922 ( CB4856>N2 ) kyIR20 II: QX111 , a RIAIL containing the HW II-QTL was backcrossed to clr-1 dpy-10 ( in an N2 background ) for 9 generations , picking non-Clr , non-Dpy males each generation . The introgression breakpoints are , on the left , between 4 , 783 , 398 ( indel ) and 4 , 800 , 876 ( marker haw25011 ) , and on the right , between 6 , 627 , 080 ( marker haw25929 ) and 6 , 672 , 356 ( marker haw25938 ) . CX13072 ( CB4856>N2 ) kyIR82 II: CX11922 kyIR20 was crossed to N2 and an F2 recombinant was made homozygous . The introgression breakpoints are , on the left , between 4 , 783 , 398 ( indel ) and 4 , 800 , 876 ( marker haw25011 ) , and on the right , between 6 , 215 , 940 ( marker haw25805 ) and 6 , 442 , 763 ( indel ) . CX13073 ( CB4856>N2 ) kyIR83 II: CX11922 kyIR20 was crossed to N2 and an F2 recombinant was made homozygous . The introgression breakpoints are , on the left , between 5 , 926 , 596 ( indel ) and 5 , 941 , 581 ( indel ) , and on the right , between 6 , 627 , 080 ( marker haw25929 ) and 6 , 672 , 356 ( marker haw25938 ) . CX13602 ( CB4856>N2 ) kyIR97 II: CX13073 kyIR83 was crossed to N2 and an F2 recombinant was made homozygous . The introgression breakpoints are , on the left , between 5 , 926 , 596 ( indel ) and 5 , 941 , 581 ( indel ) , and on the right , between 6 , 195 , 603 ( marker haw25802 ) and 6 , 198 , 696 ( marker haw25803 ) . CX13601 ( CB4856>N2 ) kyIR96 II: CX13073 kyIR83 was crossed to N2 and an F2 recombinant was made homozygous . The introgression breakpoints are , on the left , between 5 , 926 , 596 ( indel ) and 5 , 941 , 581 ( indel ) , and on the right , between 6 , 047 , 397 ( marker haw25707 ) and 6 , 064 , 898 ( marker haw25718 ) . N2 animals were crossed to CX13602 kyIR97 , F1 hermaphrodites were selfed and 5 , 000 individual F2 hermaphrodites were dispensed into single wells of 96-well plates with the use of a worm sorter ( COPAS Biosort system; Union Biometrica ) . These F2s were grown on 200 µL of an E . coli OP50 suspension in S-basal buffer with cholesterol , rotating at 230 RPM , at 22°C for 6 days ( 1–2 generations ) . The progeny of each F2 were genotyped at markers 5 , 941 , 581 ( indel ) and 6 , 195 , 603 ( marker haw25802 ) , a 0 . 41 cM interval , and recombinants between these markers were identified . Animals homozygous for the recombinant chromosome were then tested for social behavior . CX13854 ( CB4856>N2 ) kyIR102 II: The left breakpoint fell at 6 , 155 , 430 ( marker haw25773 ) ; the breakpoint on the right is between 6 , 195 , 603 ( marker haw25802 ) and 6 , 198 , 696 ( marker haw25803 ) . CX14013 ( CB4856>N2 ) kyIR112 II: The introgression breakpoints are , on the left , between 5 , 926 , 596 ( indel ) and 5 , 941 , 581 ( indel ) , and on the right , between 6 , 132 , 624 ( marker haw25748 ) and 6 , 146 , 765 ( marker haw25767 ) . CX14008 ( CB4856>N2 ) kyIR107 II: The introgression breakpoints are , on the left , between 5 , 926 , 596 ( indel ) and 5 , 941 , 581 ( indel ) , and on the right , between 6 , 147 , 008 ( marker haw25768 ) and 6 , 148 , 900 ( marker haw25769 ) . CX13845 ( CB4856>N2 ) kyIR98 II: The introgression breakpoints are , on the left , between 5 , 926 , 596 ( indel ) and 5 , 941 , 581 ( indel ) , and on the right , between 6 , 156 , 160 ( marker haw25774 ) and 6 , 156 , 620 ( marker haw25775 ) . CX14011 ( CB4856>N2 ) kyIR110 II: The introgression breakpoints are , on the left , between 6 , 111 , 057 ( marker haw25735 ) and 6 , 117 , 083 ( marker haw25738 ) , and on the right , between 6 , 156 , 160 ( marker haw25774 ) and 6 , 156 , 620 ( marker haw25775 ) . The abts-3d isoform was not present in the standard Gene Models of Wormbase ( WS229 ) . The existence of an alternative first exon in abts-3d was inferred from the following evidence: ( 1 ) presence of Illumina sequence reads from cDNA derived from polyA+ RNA of larval and adult C . elegans covering the putative exon , including reads that span an inferred 3′ exon junction but absence of 5′ exon junction reads ( as shown in Wormbase ) , ( 2 ) presence of a long open reading frame that overlaps the cDNA reads , ( 3 ) high degree of conservation of the putative exon with other Caenorhabditis species , relative to other introns in the gene , ( 4 ) high evolution rate across Caenorhabditis species of third positions of putative codons in the region relative to first and second positions , ( 5 ) presence of a 3-bp indel in the region , an unusual feature in non-coding sequences . Animals were synchronized in the L1 stage ( 16 to 20 h post-egg laying ) by allowing adults to lay eggs on seeded plates for four hours . The L1 stage was chosen because ( 1 ) it is an important time point for daf-7 regulation , ( 2 ) neuronal genes are expressed at the highest relative level with respect to total RNA in L1 animals , and ( 3 ) it is easy to maintain tight developmental synchrony at this stage . Total RNA from these L1 stage synchronized cultures was isolated with Trizol-chloroform , precipitated with an equal volume of 70% ethanol and cleaned with Zymo Quick-RNA MicroPrep according to the manufacturer's instuructions . 800 ng of RNA and oligo-dT were used for reverse transcription using SuperScript III First-Strand Synthesis ( Invitrogen ) according to the manufacturer's instructions . Real-time PCR was performed with Fast SYBR Green Master Mix ( Applied Biosystems ) on a 7900HT Real-Time PCR System ( Applied Biosystems ) . cdc-42 was used as the calibrator for relative quantitation [64] . Primers used were: exp-1_F , ttttggcagatttcaacagc exp-1_R , ttcatcattttcctccatcaag daf-7_F , gcaccaactcaggtgtttgtat daf-7_R , aatccctttggtgcctcttt cdc-42_F , cggatgttggagagaagttgg cdc-42_R , ctgttgtggtgggtcgagag Transgenes were made by injection of DNA clones into the gonads of young adult hermaphrodites together with a fluorescent coinjection marker [65] . To control for variation among transgenes , at least three independent lines from each injection were characterized . The following fosmids were injected alone or in combination , as described below in Transgenic strains: WRM066cE09 , WRM0620bF02 , WRM0610bG09 , H35N03 , WRM0612dD07 . Prior to injection , fosmid structures were confirmed by restriction digest with diagnostic enzymes . pAB05 is a 7 . 6 kb genomic exp-1 NsiI/ScaI fragment that contains an in-frame GFP fusion in the intracellular loop between M3 and M4 of EXP-1 [40] . | In both animals and humans , normal individuals can behave differently in the same environment . Natural variation in behavior is partly due to genetic differences between individuals and partly due to experience . Mapping studies have demonstrated that the genetic component of natural behavioral variation is complex , with many genes that each contribute a small amount to the observed behavior . This complexity has made it difficult to identify the causative genes for individual differences . Here we use the nematode worm C . elegans to dissect a social behavioral trait , the propensity to aggregate with other animals in the presence of food . We find that the behavioral differences between two wild-type worm strains result from at least five genetic differences between the strains , two of which were previously known . One of the three new loci affects a receptor for the neurotransmitter GABA , which regulates excitability in the brain . In the context of previous work , we suggest that a significant number of genes that generate behavioral variation encode neurotransmitter receptors . This analysis in a model animal may help guide discoveries of the genetic variants that affect common human behavioral traits by suggesting classes of genes to examine closely . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] | [
"animal",
"models",
"caenorhabditis",
"elegans",
"behavioral",
"neuroscience",
"model",
"organisms",
"genetic",
"polymorphism",
"genetics",
"population",
"genetics",
"biology",
"neurotransmitters",
"neuroethology",
"neuroscience",
"genetics",
"and",
"genomics",
"gene",
"function"
] | 2012 | Long-Range Regulatory Polymorphisms Affecting a GABA Receptor Constitute a Quantitative Trait Locus (QTL) for Social Behavior in Caenorhabditis elegans |
The Endosomal Sorting Complexes Required for Transport ( ESCRT ) machinery , a highly conserved set of four hetero-oligomeric protein complexes , is required for multivesicular body formation , sorting ubiquitinylated membrane proteins for lysosomal degradation , cytokinesis and the final stages of assembly of a number of enveloped viruses , including the human immunodeficiency viruses . Here , we show an additional role for the ESCRT machinery in HIV-1 release . BST-2/tetherin is a restriction factor that impedes HIV release by tethering mature virus particles to the plasma membrane . We found that HRS , a key component of the ESCRT-0 complex , promotes efficient release of HIV-1 and that siRNA-mediated HRS depletion induces a BST-2/tetherin phenotype . This activity is related to the ability of the HIV-1 Vpu protein to down-regulate BST-2/tetherin . We found that BST-2/tetherin undergoes constitutive ESCRT-dependent sorting for lysosomal degradation and that this degradation is enhanced by Vpu expression . We demonstrate that Vpu-mediated BST-2/tetherin down-modulation and degradation require HRS ( ESCRT-0 ) function and that knock down of HRS increases cellular levels of BST-2/tetherin and restricts virus release . Furthermore , HRS co-precipitates with Vpu and BST-2 . Our results provide further insight into the mechanism by which Vpu counteracts BST-2/tetherin and promotes HIV-1 dissemination , and they highlight an additional role for the ESCRT machinery in virus release .
The assembly and release of HIV-1 particles requires a highly orchestrated series of interactions between proteins encoded by the virus and key cellular components , including elements of the cellular membrane trafficking apparatus and the ESCRT ( Endosomal Sorting Complexes Required for Transport ) machinery [1] , [2] , [3] . The ESCRT machinery was initially found to be involved in the recognition and sorting of membrane proteins to the internal vesicles of multivesicular bodies ( MVBs ) /late endosomes and the subsequent degradation of these cargoes in lysosomes . Subsequently , the ESCRT machinery was found to also play key roles in topologically related membrane scission reactions that occur during cytokinesis and the budding of a number of enveloped viruses , including HIV-1 [1] , [4] . The ESCRT machinery comprises four multiprotein complexes ESCRT-0 , -I , -II and -III whose sequential recruitment to endosomal membranes mediates ( 1 ) the formation ( budding ) and subsequent pinching off of intralumenal endosomal vesicles into MVBs and ( 2 ) the incorporation of ubiquitinylated membrane protein cargoes into the inwardly budding vesicles [4] . The ESCRT-0 protein HRS ( also called hepatocyte growth factor-regulated tyrosine kinase substrate [HGS] ) initiates this process by acting as a linker protein that binds , on the one hand , ubiquitinylated cargoes and , on the other , a PSAP motif in the ESCRT-I component TSG101 [4] . Topologically , the budding of HIV-1 particles at the plasma membrane resembles the budding of intralumenal vesicles into MVBs . Moreover , the HIV-1 Gag protein , the major structural protein of the virus , can recruit the ESCRT-I and ESCRT-III complexes through its C-terminal p6 domain ( so called “late domain” ) to mediate release of budding virions [5] , [6] , [7] , [8] . A PTAP motif in the Gag p6 domain binds TSG101 , mimicking the HRS-PSAP mediated recruitment of ESCRT-I . Failure to recruit ESCRT-I , and consequently ESCRT-III , leads to accumulation of viral budding intermediates at the surface of infected cells [5] , [8] . In addition , a LYPXnL motif in the Gag p6 domain provides a second means to recruit the ESCRT machinery by binding the ESCRT-I and ESCRT-III interacting protein ALIX/AIP1 [7] . Although Gag mimics HRS in binding TSG101 , HRS itself is not required for the HIV-1 budding and scission reactions [9] . Nevertheless , HRS was recently identified in a genome-wide siRNA screen for host cell factors involved in HIV-1 replication , suggesting an active role for ESCRT-0 [10] . In this study , we explored the role of HRS in the HIV-1 replication cycle . Using RNA interference and virological assays , we show that HRS is required for efficient release of HIV-1 particles . HRS function was not related to the described TSG101 activity in HIV-1 scission , but was related to the ability of the accessory viral protein Vpu to down-regulate the protein bone marrow stromal antigen 2 ( BST-2 also called CD317/HM1 . 24/tetherin ) . BST-2 was recently identified as a cellular restriction factor that impedes the release of fully assembled HIV-1 by physically tethering particles to the plasma membrane of infected cells . Vpu counteracts this restriction and induces the down-regulation of BST-2 expressed at the cell surface . Vpu-induced cell surface down-regulation of BST-2 is associated with targeting of the cellular protein for degradation [11] , [12] , [13] , [14] , [15] . Here we show that BST-2 undergoes constitutive ESCRT-mediated lysosomal degradation . Moreover , HRS facilitates Vpu-induced BST-2 down-regulation and degradation , and therefore contributes to efficient HIV-1 release . Furthermore HRS co-precipitates with Vpu and BST-2 . Altogether , our results highlight an additional role for the ESCRT machinery in HIV-1 release , and provide further understanding of the mechanism by which Vpu targets BST-2 for degradation .
To investigate the role of the ESCRT-0 component HRS on HIV-1 replication , we specifically depleted HRS in HeLa cells and analysed the effect on HIV-1 propagation . We used a siRNA against HRS that has been described and validated in earlier studies ( [16] Cf . Experimental procedures ) . By western-blot analysis , siRNA transfections into HeLa cells resulted in >95% reduction in HRS levels , as previously described ( [16]; Figures 1A and C ) . We assessed the impact of HRS silencing on the ability of the NL4-3 strain of HIV-1 ( NL4-3 WT ) to propagate in the reporter cell line HeLa P4R5 ( stably expressing the receptor CD4 and the coreceptor CCR5 ) . The cells were either treated with a control siRNA ( siRNA CT; referred to as control cells ) or with siRNA targeting HRS ( siRNA HRS ) , then infected with NL4-3 HIV-1 at a MOI of 0 . 005 and left for 4 days to allow the virus to propagate through the cultures . HIV-1 production was scored by ELISA quantification of the Gag product CAp24 , reflecting the amount of virus present in the supernatant . Figure 1B shows that depletion of HRS caused a significant reduction in virus release , with <10% CAp24 detected in the supernatant of these cells compared to the control cells , suggesting that HRS is required for optimal HIV-1 replication . As HIV-1 entry into HRS-depleted cells was not affected ( Figure S1 ) , we tested whether the reduction in HIV-1 propagation observed following HRS knock down was a consequence of decreased production of viral particles . To this end , HeLa cells were transfected with control siRNA ( siRNA CT ) or siRNA targeting HRS ( siRNA HRS ) , and then infected with NL4-3 HIV-1 pseudotyped with VSV-G ( NL4-3 WT ) at an MOI = 0 . 5 . VSV-G pseudotyping enables HIV-1 production to be monitored after a single round of infection . Forty-eight hours after infection the protein content of the cells was analyzed by western-blot ( Figure 1C ) and the virus released into the supernatant was assessed by western-blot and ELISA quantification of CAp24 ( Figures 1D–E ) . Depletion of HRS ( Figure 1C ) did not significantly modify the maturation of Gag compared to the control cells . However , in HRS-depleted cells a clear accumulation of Env ( gp160 , SUgp120 and TMgp41 ) and of the processed Gag products , CAp24 and MAp17 , was observed ( Figures 1C–E ) . The expression levels of the viral accessory proteins Nef and Vpu were similar in both CT and HRS knock down cells . As these accessory proteins are not incorporated significantly into nascent viral particles , we postulated that the increased level of cell-associated Env and processed Gag products , CAp24 and MAp17 , might be due to a decrease in HIV-1 release from HRS depleted cells . Quantification of the amount of CAp24 in the cell supernatants ( Figures 1D and 1E ) and western-blot analysis of viral preparations ( Figure 1C , lower panel ) confirmed that virus release was decreased by 60–70% in HRS knock down cells . These data indicate that HRS is required for efficient HIV-1 release from HeLa cells . To understand the mechanism by which HRS contributes to HIV-1 release , we assessed whether HRS depletion perturbed Gag recruitment of TSG101 . To this end , we analysed the ability of a GST-p6 fusion construct , corresponding to the sub-domain of Gag bearing the PTAP motif involved in TSG101 interaction , to interact with TSG101 in control and HRS depleted cells . Figure 2A shows similar levels of binding of GST-p6 to TSG101 in control and HRS depleted cells ( Figure 2A , compare lane 7 with lane 8 ) . Similar results were obtained using a GST-Gag construct ( Figure S2 ) suggesting that HRS is not required for p6-mediated binding of Gag to TSG101 and recruitment of ESCRT-I . We then analyzed whether depletion of HRS and TSG101 led to similar alteration of virus production using the assay described above . As shown in Figure 2C , both HRS and TSG101 depletion reduced HIV-1 release from infected cells , as assessed by quantitative ELISA . However , western-blot analysis of cell-associated viral proteins suggested that HRS and TSG101 have different roles in HIV-1 production ( Figure 2B ) . Depletion of TSG101 , as well as expression of viruses mutated in the Gag PTAP motif , induced a profound alteration in Gag processing , characterized by an accumulation of the p41 and CA-SP1 Gag processing intermediates as previously described [5] , [6] . No such alteration was observed upon depletion of HRS ( Figures 1C and 2B ) . Infectious viral release from the siRNA-transfected cells was then scored using the indicator cell line HeLa TZM-bl , and the results were normalized to the amount of CAp24 present in the cell supernatants . As previously shown , particles produced by TSG101 depleted cells were mostly non-infectious due to defects in Gag processing ( 80% loss of infectivity compared to virions released from control cells ) [5] . By contrast , the virions produced from HRS depleted cells exhibited <45% loss of infectivity compared to control viruses ( Figure 2D ) . Interestingly , western-blot analysis of viral particles showed a decrease in envelope glycoprotein ( Env ) incorporation ( Figure S3 ) , which may explain this lower infectivity . HRS might also be required for the proper intracellular trafficking and processing of Env within infected cells . Therefore , HRS depletion might result in intracellular accumulation and incorporation of mature and unprocessed Env products unfit for HIV-1 infectivity ( Figures 1C , 2B , 2D ) . Taken together , our data indicate that HRS is required for efficient HIV-1 release from HeLa cells , and its mode of action is distinct from that of TSG101 in HIV-1 budding . To further explore the mechanism by which HRS reduces HIV-1 release , siRNA treated HeLa cells infected with VSV-G pseudotyped NL4-3 HIV-1 were prepared for cryosectioning and electron microscopy ( EM ) . Initially , immunofluorescence staining of semi-thin survey cryosections with antibodies against the viral proteins MAp17 and Env revealed a large number of HIV-infected cells ( Figure 3A and 3B ) . The anti-MAp17 antibody 4C9 , which detects only the cleaved Gag product MA , labelled small spots probably representing mature virions , as well as larger clumps of viruses , confirming that HRS depletion does not inhibit Gag processing and maturation of HIV-1 particles . Staining with antibodies against Env was observed over the endoplasmic reticulum , with juxta-nuclear patches of brighter staining , and some Env labelling co-localized with p17 stained viruses . On sections from HRS knockdown cells , the staining for p17 and Env was much stronger than on the control cells , with large clumps of viruses observed near the cell surface , sometimes attached via fine processes . Some virus was also observed in cytoplasmic vacuoles ( Figure 3B ) . To analyse the accumulations of virus particles in more detail , we stained ultrathin cryosections ( 50 nm ) for p24/p55Gag and Env with protein A-gold ( PAG ) for EM . For both the control and HRS siRNA-treated samples , infected cells could be identified by scattered p55Gag labelling ( 5 nm PAG ) over the cytoplasm , but not over nuclei , demonstrating specificity of the labelling ( Figures 3C–E and 3H ) . Env was seen over the Golgi apparatus and associated small tubulo-vesicular membranes ( Figure S4A and S4C ) . Virus particles were labelled with antibodies against both p24/p55Gag ( 5 nm PAG ) and Env ( 10 nm PAG ) , showing that Env-containing virions were produced by control and HRS-depleted cells . On control cells , the virus particles were seen either singly or in small clusters at the cell surface ( Figure 3C and 3D ) , and occasionally virus particles could also be observed in intracellular vacuoles ( Figure 3E ) . On the HRS-depleted cells , much larger clusters of HIV particles were attached to the cell surface either directly or via microvillar protrusions ( Figure 3F , the upper of these virus clusters is enlarged in Figure 3G ) . Some of these clusters contained >20 and sometimes as many as 80–100 virus particles . In addition , we observed intracellular vacuoles resembling endosomes containing many p24/p55Gag and Env-immunolabelled virus particles in some of the HRS-depleted cells ( Figure 3H , 3I and S4F ) . We confirmed these observations by counting the distribution of 669 and 395 virus particles in the HRS-depleted and control cells , respectively . On the HRS-depleted cells , 47% of the virus particles were in clusters of more than 20 particles , while in control cells about half of the viruses occurred alone or in groups of 2–4 particles , and no virus clusters contained more than 18 virions . In addition , for the HRS-depleted cells , 27% of the virus particles were seen in intracellular vesicles , while for the control cells only 7% of virus particles were in endosomes . The large extracellular virus clusters and accumulation of virus particles in endosomes in the HRS-depleted cells was reminiscent of the phenotype observed in BST-2/tetherin-expressing cells infected with Vpu-defective HIV-1 strains [13] , [15] . This suggests that depletion of HRS in HeLa cells leads to tethering of viral particles at the cell surface and their subsequent endocytosis , and that HRS knock down may affect the trafficking of BST-2 . The HIV-1 accessory protein Vpu counteracts the cellular restriction factor BST-2 to promote HIV-1 release . BST-2 physically tethers newly formed viral particles at the surface of infected cells , preventing their release . Vpu relieves this restriction by down regulating cell surface expression of BST-2 . In addition , Vpu promotes the degradation of BST-2 [11] , [12] , [13] , [14] , [15] , [17] . Based on the EM observations above , we postulated that HRS might contribute to the process of Vpu inhibition of BST-2 restriction . As the ESCRT machinery sorts ubiquitinylated membrane proteins for lysosomal degradation , we analysed whether HRS , and by extension the MVB pathway , participates in the degradation of BST-2 . HeLa cells were transfected with siRNAs targeting HRS or TSG101 and the turnover of BST-2 was monitored after incubating the cells in growth medium containing cycloheximide . We observed that almost 90% of BST-2 was degraded in 4 hours in cells transfected with control siRNA ( Figure 4A; siRNA CT ) . By contrast , in HRS depleted cells , the half-life of BST-2 was prolonged , with only 30% of the initial BST-2 pool degraded after 4 hours ( Figure 4A; siRNA HRS ) . Western-blot and immunofluorescence analysis revealed increased BST-2 expression in HRS depleted cells compared to the control cells , consistent with stabilization of the protein ( Figure S5A and C ) . We also investigated the role of TSG101 in the degradation of BST-2 . As TSG101 depletion is achieved more rapidly than HRS depletion ( ∼2 days and 4 days , respectively ) , the cells were harvested 72 hours after siRNA transfection ( Figure 4B ) . As observed with HRS knock down , loss of TSG101 also stabilized BST-2 expression ( less than 10% degradation in 4 hrs; Figure 4B ) . This stabilization was also associated with a striking accumulation of BST-2 within intracellular structures that partially co-labelled for HRS ( Figure S5B and C ) . As an internal control , we monitored the turnover of EGF receptors ( EGF-R ) following EGF stimulation . Upon ligand binding , EGF-R is targeted to MVBs prior to lysosomal degradation [18] , [19] . Consistent with these previous studies , knock down of HRS ( Figure 4A ) or TSG101 ( Figure 4B ) inhibited EGF-induced EGF-R degradation compared to control cells ( Figures 4A–B , lower panels ) . These data show that HRS and TSG101 are involved in the sorting of BST-2 for degradation in the MVB/lysosomal pathway . To obtain further evidence for the involvement of the ESCRT machinery in BST-2 trafficking , we analysed the effect of over-expressing HRS , or expressing a dominant negative mutant of the v-ATPase VPS4 ( VPS4-E223/Q ) , on BST-2 expression . Over-expression of HRS was shown to negatively perturb the ESCRT pathway [18] . VPS4 is an AAA-ATPase involved in recycling the ESCRT machinery by facilitating its dissociation from endosomal membranes [20] . Mutation of E223 to Q blocks ATP hydrolysis and renders VPS4-E223/Q a dominant-negative inhibitor of the ESCRT pathway [20] . Immunofluorescence microscopy showed that in cells expressing GFP-HRS , BST-2 accumulated within enlarged endosomal structures that co-labelled with GFP-HRS , compared to the neighbouring cells in which BST-2 is expressed in discrete punctuate structures ( Figure 4C , panels d-f ) . Similarly , expression of the VPS4 dominant negative mutant resulted in a striking accumulation of BST-2 ( Figure 4C , panels j-l ) , suggesting that expression of VPS4-E223/Q as well as over-expression of HRS inhibited BST-2 degradation compared to control cells expressing wild type VPS4 or GFP alone ( Figure 4C ) . Together , our results indicate that BST-2 is rapidly turned over in HeLa cells and that the ESCRT machinery is required for its efficient degradation . As Vpu expression promotes the release of virus particles from restrictive BST-2-positive cells [13] , [15] , we analysed the effect of HRS depletion on the production of Vpu-defective HIV-1 . To this end , HeLa cells were transfected with control siRNA or siRNA targeting HRS and then infected with either VSV-G pseudotyped wt NL4-3 HIV-1 ( NL4-3 WT ) or VSV-G pseudotyped Vpu-defective NL4-3 HIV-1 ( NL4-3 Udel ) at an equivalent MOI ( MOI = 0 . 5 ) . The cell proteins were then analysed by western-blot and the viral particles released into the cell supernatants measured by CAp24 ELISA and analysed by western-blot . As previously described [13] , [15] , the release of Vpu-defective HIV-1 particles from restrictive cells was poor compared to wild type virus ( Figure 5 ) . Interestingly , in HRS knock down cells , the release of Vpu-deleted HIV-1 was also lower ( ∼40% less virus ) compared to the control cells ( Figure 5A , lower panel and Figure 5B ) . As observed for WT viruses , the reduced release of Vpu-defective HIV-1 observed in HRS depleted cells was associated with increased amounts of cell-associated Gag/CAp24 ( Figure 5B ) . These results suggest cooperation between Vpu and HRS to promote efficient HIV-1 release . Cells such as PBMC , HEK 293T , COS , HT1080 , and HOS do not normally express BST-2 . These cells are referred to as ‘non-restrictive’ and release the Vpu-defective HIV-1 as efficiently as wild type viruses [13] , [15] , [21] . To further explore the role of HRS in the BST-2 restriction of HIV-1 release , we compared the effect of HRS depletion on HIV-1 production in non-restrictive HEK 293T cells and restrictive HeLa cells ( Figure 6 ) . RNAi mediated HRS knock down was similar in HEK 293T and HeLa cells ( >95%; Figure 6 ) . In HeLa cells ( Figure 6A and 6B ) , HRS depletion inhibited the release of HIV-1 and induced an accumulation of cell-associated CAp24 as shown before . In the HEK 293T cells , a reduced amount of CAp24 was detected in the supernatant , but this was paralleled by a decrease in the amount of cell-associated CAp24 ( Figure 6A and 6C ) compared to the control . Thus , in HEK 293T cells the decreased p24 release most likely reflects reduced HIV-1 infection rather than a defect in virus release ( Figure 6C ) . To circumvent the caveats caused by the lower sensitivity of HRS depleted HEK 293T cells to HIV-1 infection , we analysed the outcome of HRS depletion in HeLa cells depleted for BST-2 ( Figure 7 ) . HeLa cells were transfected with CT siRNA or siRNA targeting HRS and BST-2 alone or together and then infected with either VSV-G pseudotyped wt NL4-3 HIV-1 ( NL4-3 WT ) or VSV-G pseudotyped Vpu-defective NL4-3 HIV-1 ( NL4-3 Udel ) at an equivalent MOI ( MOI = 0 . 5 ) . As previously described , the release of Vpu-defective HIV-1 was rescued in cells depleted for BST-2 compared to the control cells ( Figure 7A lower panel; compare lane 7 to lane 5 , see Figure 7B ) , consistent with a restriction of Vpu-defective virus release by BST-2 [13] , [15] . Interestingly HRS depletion did not affect the release of NL4-3 wt viral particles in BST-2 depleted cells ( lower panel , compare lane 4 to lane 2; Figure 7B ) , as similar amounts of virus were produced from these cells compared to the control cells ( compare lane 4 to lane 1 ) . Similarly , release of NL4-3 Udel viruses was no longer affected by HRS depletion in HeLa cells devoid of BST-2 ( Figure 7 , lower panel , compare lane 8 to lane 6 , see Figure 7B ) . These results support the notion that HRS does not play a major role in HIV-1 release from non-restrictive cells and suggest that HRS may contribute to the activity of Vpu in counteracting BST-2/tetherin mediated restriction of virus release . We observed that HRS is involved in the constitutive degradation of BST-2 ( Figure 4A ) and appears to promote HIV-1 release from restrictive cells ( Figures 1C–E ) . We therefore tested whether HRS is required for Vpu induced BST-2 degradation . RNAi-treated HeLa cells were infected with either VSV-G pseudotyped wt NL4-3 ( NL4-3 WT ) or Vpu-deleted NL4-3 ( NL4-3 Udel ) viruses at a MOI of 1 to infect the majority of the cells . Forty-eight hours after infection , the cells were incubated in medium containing cycloheximide to monitor the degradation of BST-2 with time ( Figure 8A ) . As described above ( Figure 4 ) , the level of BST-2 decreased in control cells in 4 hours of treatment with cycloheximide . By contrast , BST-2 levels were stabilized in HRS depleted cells over 4 hours ( Figure 8A , compare lanes 1–2 with lanes 3–4 ) . A similar pattern was observed in cells infected with Vpu-defective viruses ( lanes 9–10 to 11–12 ) . In control cells infected with NL4-3 WT , BST-2 was barely detectable at time 0 , consistent with Vpu-enhanced degradation of BST-2 ( lanes 5–6 ) . Interestingly , in HRS depleted cells , BST-2 levels were similar to those in uninfected cells or cells infected with Vpu-defective viruses at time 0 and remained stable during 4 h of treatment with cycloheximide ( Figure 8A , lanes 7–8 ) . Immunofluorescence staining showed that , in control cells , HIV-1 infection ( revealed by staining for Env ) led to reduced BST-2 staining , compared to neighbouring non-infected cells , whereas BST-2 labelling was comparable in infected and non-infected cells when HRS was depleted ( Figure 8B ) . These data show that HRS depletion impedes the ability of Vpu to target BST-2 for degradation . Taken together , our data indicate that Vpu-induced BST-2 degradation involves the ESCRT/MVB pathway . Although Vpu expression can induce BST-2 degradation , the mechanism by which Vpu inhibits BST-2 restriction of HIV-1 release is unclear . Treatment of cells with the v-ATPase inhibitor bafilomycin A1 , which inhibits both endosomal sorting and lysosomal function , can restore BST-2 cell surface levels in cells expressing Vpu , most likely due to the recycling of BST-2 [17] . We thus wondered whether depleting HRS counteracts Vpu-mediated down regulation of cell surface BST-2 . To answer this question , siRNA transfected HeLa cells were infected with VSV-G pseudotyped wt NL4-3 HIV-1 ( NL4-3 WT ) at a MOI of 0 . 5 so that approximately 50% of the cells were infected . Cell surface levels of BST-2 were assessed by flow cytometry ( Figures 8C–D and Table S1 ) . Intracellular staining of CAp24 ( CAp24-FITC ) was used to distinguish non-infected cells ( Figure 8C–D: gate B1 and black bars ) from infected cells ( Figure 8C–D: gate B2 and grey bars ) . In control cells ( siRNA CT ) infected with WT NL4-3 HIV-1 , BST-2 cell surface expression decreased by ≥60% compared with non-infected cells , consistent with previous studies [15] , [22] . Interestingly , the Vpu-induced down regulation of cell surface BST-2 was impaired upon depletion of HRS since no significant decrease of cell surface BST-2 was observed on infected HRS depleted cells ( less than 20% decrease ) compared to non-infected HRS depleted cells ( Figure 8C–D and Table S1 ) . Our data suggest that disruption of BST-2 sorting in endosomes by HRS depletion diminishes the ability of Vpu to down-regulate cell surface BST-2 , most likely by allowing internalized BST-2 to recycle to the cell surface . To obtain further evidence that the mechanism by which Vpu counteracts BST-2 involves the ESCRT machinery , we analysed the effect of over-expressing a dominant negative mutant of the AAA-ATPase VPS4 ( VPS4-E223/Q ) on Vpu-induced down-regulation of cell surface BST-2 . HeLa cells were co-transfected with plasmids encoding wt VPS4 or VPS4-E223/Q fused to GFP along with the plasmid encoding the provirus wt NL4-3 HIV-1 ( NL4-3 wt ) or Vpu-defective NL4-3 HIV-1 ( NL4-3 Udel ) as a control . Cell surface levels of BST-2 were assessed by flow cytometry on GFP positive infected cells ( surface staining of Env was used to identify HIV-1 expressing cells ) ( Figure 9 ) . In cells expressing wt VPS4 , a marked decrease of cell surface BST-2 was observed in cells expressing HIV-1 NL4-3 wt compared to cells expressing Vpu-defective viruses ( Figure 9A , upper left panel and Figure 9B ) , consistent with Vpu-induced down–regulation of cell surface BST-2 [11] , [12] , [15] , [17] , [22] . Interestingly , expression of VPS4-E223/Q impaired the ability of Vpu to down-regulate BST-2 from the cell surface , as similar levels of BST-2 were detected at the surface of cells expressing WT and Vpu-defective viruses ( Figure 9 , upper right panel and Figure 9B ) . Together with the results obtained from depleting HRS ( Figure 8C–D ) , our data suggest that the integrity of the ESCRT machinery is required for Vpu to down-regulate BST-2 from the cell surface . We showed that HRS is required for the targeting of BST-2 to the degradation pathway ( Figure 4 ) and for the ability of Vpu to down-regulate BST-2 ( Figures 5 and 8 ) to promote efficient release of HIV-1 particles . Since HRS is physiologically involved in the recognition of cargo on endosomal membranes , we postulated that HRS might interact with BST-2 and/or Vpu . To test this , HEK 293T cells were transfected with plasmids encoding HRS fused to SBP-CBP tags ( SBP-CBP-HRS ) along with Flag-BST-2 and/or HA-Vpu . Interaction with HRS was assessed by pull-down of SBP-CBP-HRS followed by western-blot analyses of bound proteins . Figure 10 shows that HRS interacts with Flag-BST-2 ( Figure 10 , Lane 6 ) and is also able to bind HA-Vpu ( Figure 10 , Lane 7 ) . Interestingly all three proteins co-precipitated when expressed together ( Figure 10 , Lane 8 ) , and HRS apparently bound more BST-2 in the presence of Vpu . This suggests that Vpu enhances the recognition of BST-2 by the ESCRT machinery , perhaps by bridging BST-2 and HRS .
Studies of the mechanism through which HIV-1 particles are assembled and released from infected cells have revealed a major role for the ESCRT machinery . ESCRT-I and ESCRT-III complexes are recruited to sites of virus assembly by binding to the viral Gag protein , promoting the scission of assembled viral particles [5] , [6] , [7] , [8] . An additional step in the release of mature virus has been highlighted through studies of the HIV-1 accessory protein Vpu . Vpu had been implicated in the efficient release of HIV-1 from certain cell lines ( referred to as restrictive cells ) but its cellular target was unknown . Recently , BST-2/tetherin was identified as a cellular restriction factor that impedes the release of HIV-1 by physically tethering fully formed mature particles to the plasma membrane of infected cells . It was proposed that , by reducing BST-2/tetherin cell surface expression , Vpu abrogates BST-2/tetherin function , thus allowing efficient virus release [12] , [13] , [14] , [15] , [17] , [22] . Here we show that the HRS component of the ESCRT-0 complex facilitates Vpu-induced BST-2 down-regulation and degradation , thereby contributing to efficient HIV-1 release . Our data thus highlight an additional role for the ESCRT machinery in HIV-1 release , following ESCRT-mediated membrane scission . Physiologically , HRS is thought to initiate the ESCRT-mediated formation of MVBs . A PSAP motif in HRS recruits the ESCRT machinery to endosomal membranes through interaction with the ESCRT-I component TSG101 [16] . These events are mimicked by HIV when ESCRT-I is recruited to budding sites through a similar PTAP motif in the C-terminal p6 “late domain” of Gag [6] , [9] . Although HRS is not needed to recruit the ESCRT machinery for viral scission [9] , we show that it is required to promote efficient Vpu-dependent release of HIV-1 particles following completion of the budding reactions . The reduced HIV-1 release caused by HRS depletion is not a consequence of an indirect effect on the ability of Gag to recruit TSG101 , as HRS depletion neither impaired the binding of Gag to TSG101 nor altered Gag processing ( Figure 2; [5] ) . A role for HRS in HIV-1 morphogenesis was previously investigated in HEK 293T cells . In that study , over-expression of HRS or HRS deletion mutants induced a “late budding phenotype” by depleting the pool of free TSG101 available to Gag to promote viral scission [23] . These data are not incompatible with our current results , but use of RNA interference and BST-2 expressing cells ( HEK 293T cells do not constitutively express BST-2/tetherin and are non-restrictive for Vpu-defective HIV-1 ) allowed us to uncover an additional level of ESCRT-mediated regulation of HIV-1 release . A key finding suggesting that HRS might influence virus release was the observation of large clusters of mature virus particles on the surface and in endosomes of HIV-1 infected HRS depleted cells ( Figure 3 ) . This phenotype was reminiscent of that observed following infection of BST-2-expressing cells with Vpu-defective HIV-1 [13] and was clearly different from that induced by the absence of TSG101 recruitment to budding sites , which is characterized by the cell surface accumulation of incomplete immature viral buds [5] . This suggested that HRS might contribute to the mechanism counteracting BST-2 restriction of HIV-1 release . We therefore investigated whether the ESCRT machinery , and in particular HRS and TSG101 , facilitate BST-2 down-modulation and degradation . Previous studies have indicated that BST-2 undergoes clathrin-mediated endocytosis from the plasma membrane and traffics through recycling endosomes and the trans Golgi network ( TGN ) [24] . Indeed a significant fraction of BST-2 is located in the TGN at steady state [25] , [26] . Here we show that BST-2 undergoes rapid constitutive turnover in HeLa cells ( half time < 2 h ) that is dependent on both HRS and TSG101 , suggesting that although BST-2 cycles between the plasma membrane , endosomes and the TGN , a significant fraction is sorted to lysosomes and degraded . Interference with the ESCRT machinery , and associated effects on lysosomal sorting and degradation , by RNA interference or over-expression of WT or dominant negative proteins of the MVB pathway , resulted in inhibition of constitutive BST-2 degradation and its accumulation in intracellular structures that are likely to be endosomes and/or the TGN ( Figure 4 and Figure S5 ) . Thus , the ESCRT machinery , and in particular HRS and TSG101 , sorts BST-2 to the ESCRT/MVB degradation pathway and Vpu appears to enhance this degradation , leading to barely detectable levels of BST-2 in HIV-1 infected cells ( Figure 8 ) . Vpu abrogation of the restriction to HIV-1 release is associated with BST-2 down-regulation from the surface of infected cells [11] , [12] , [15] , [17] , [22] . To date , the exact mechanism through which Vpu reduces BST-2 expression is not clearly understood . Vpu may decrease BST-2 cell surface expression by increasing its endocytosis and/or by decreasing its recycling to the cell surface , thereby diverting more of the internalized protein to lysosomes . Recent studies reported that Vpu does not increase the rate of BST-2 endocytosis [17] and that BST-2 mutated on two cytosolic domain tyrosine residues ( Y6/Y8 ) involved in its internalization remains sensitive to Vpu [27] , [28] , [29] . However , inhibition of endocytosis by the dynamin 2-K44A mutant , as well as depletion of the clathrin adaptor AP2 , partially impair Vpu-induced BST-2 down-regulation from the plasma membrane [17] , [29] . In addition , Vpu accelerates the internalization rate of the BST-2 Y6/Y8 mutant [30] , suggesting a role for Vpu in BST-2 endocytosis . Alternatively , other studies suggest that Vpu may slow the transport of BST-2 ( both newly synthesized and recycled ) to the cell surface and sequester the restriction factor in a perinuclear compartment [27] , [31] , [32] . This last model is reminiscent of the mechanism by which the Env proteins from HIV-2 and tantalus monkey SIV ( SIVtan ) overcome BST-2 restriction , whereby BST-2 is redistributed from the cell surface to intracellular compartments , such as recycling endosomes and/or the TGN , without degradation [33] , [34] , [35] . Finally , it has been proposed that Vpu enhances the release of HIV-1 in the absence of BST-2 cell surface down-regulation and intracellular depletion [21] , [36] , suggesting another mechanism of Vpu action on BST-2 restriction . Our data indicate that the integrity of the ESCRT/MVB pathway is required for Vpu to counteract BST-2 restriction . HRS depletion reduces Vpu-induced cell surface down-regulation of BST-2 , impairs Vpu-induced BST-2 degradation and reduces HIV-1 release ( Figures 1 , 8 and S4 ) . Moreover , expression of dominant negative mutant of the AAA-ATPase VPS4 ( VPS4-E223Q ) , another component of the ESCRT machinery , impairs Vpu-induced cell surface down–regulation of BST-2 ( Figure 9 ) . Together with observations that inhibitors of the endosomal/lysosomal pathway impair Vpu-induced BST-2 down-regulation [17] , [22] , our data support a model in which Vpu may sort BST-2 at the level of endosomes , reducing its recycling to the plasma membrane [17] , [22] . This model is also in agreement with the redistribution of BST-2 to transferrin receptor positive endosomal compartments upon HIV-1 infection , as described by Habermann et al . [37] . Accordingly , one could postulate that the accumulation of BST-2 in HRS depleted cells exceeds the degradation capacity of Vpu and inhibits the release of wild-type HIV-1 by misrouting BST-2 back to the cell surface . Vpu downregulates the cell surface expression of BST-2 and increases its degradation [11] , [12] , [15] , [17] , [22] . Whether Vpu targets BST-2 for proteasomal or lysosomal degradation is still a matter of debate . Experiments using proteasome inhibitors or a dominant negative mutant of ubiquitin have suggested that Vpu connects BST-2 to the Skp1-cullin-F-Box ( SCF ) ubiquitin ligase complex and targets it for proteasomal degradation , as described for Vpu-mediated degradation of CD4 [11] , [12] , [38] . However , prolonged treatment with proteasome inhibitors can deplete cellular pools of free ubiquitin and thereby indirectly affect alternative ubiquitin-dependent sorting and degradation pathways [18] . Studies using inhibitors of the endosomal/lysosomal pathway support a model in which Vpu targets BST-2 for lysosomal degradation [17] , [22] . Although our data cannot rule out the involvement of the proteasome pathway , the involvement of HRS and the ESCRT machinery in both constitutive and Vpu-enhanced degradation of BST-2 strongly supports a role for lysosomal degradation in BST-2 down regulation . How Vpu targets BST-2 to the MVB pathway is also unclear . Studies have reported that Vpu interacts with BST-2 through the transmembrane domains of both proteins [27] , [29] and endosomal co-localisation of BST-2 with Vpu has been described [13] , [15] . Interestingly , we showed that HRS interacts with BST-2 and Vpu ( Figure 10 ) . One model to explain these observations is that through binding with BST-2 [27] , [29] and HRS , Vpu might stabilize BST-2 on endosomal membranes and , by recruiting the β-TRCP2-Skp1-cullin E3 ubiquitin ligase complex , allow the efficient ubiquitination of BST-2 to permit its recognition by the ESCRT machinery through HRS , as discussed in a recent study [17] . Indeed , recent studies have shown that β-TRCP2 , a subunit of the Skp1-cullin1-F-Box ( SCF ) ubiquitin ligase complex is required for Vpu to down-regulate BST-2 expression [12] , [17] , [22] , [29] and the SCF ubiquitin ligase complex can be recruited to endocytic membranes [39] . Moreover , it has been described that certain receptors , such as the growth hormone receptor ( GHR ) and the interferon α receptor ( IFNAR1 ) , are targeted for lysosomal degradation following their binding to β-TRCP2 and subsequent ubiquitination [40] , [41] . Physiologically , HRS is involved in the recognition of ubiquitinylated protein flagged for lysosomal degradation . BST-2 was recently shown to be ubiquitinylated and Vpu expression was shown to increase this ubiquitination [42] , [43] . BST-2 contains two lysine residues , the common target of ubiquitination , in its cytosolic N-terminal domain at positions 18 and 21 . Mutation of these residues does not impair BST-2 sensitivity to Vpu-mediated down-regulation , suggesting that BST-2 ubiquitination is not required for Vpu to target BST-2 to the MVB degradation pathway [12] , [28] , [42] . However , residues other than lysine may undergo ubiquitination ( e . g . cysteine , threonine and serine residues ) . Ubiquitination of lysine and serine/threonine residues in the cytosolic domain of CD4 has been shown to be required for its targeting to the degradation pathway by Vpu [44] . BST-2 contains a threonine and two serine residues at positions 3 to 5 and , interestingly , ubiquitination of BST-2 on these residues was recently shown to be involved in Vpu-induced release of viral particles [43] . These observations together with indirect evidence from the use of drugs that affect cellular ubiquitin pools , suggest a role for ubiquitin in Vpu-induced targeting of BST-2 to the degradation pathway [11] , [12] , [17] , [43] . Further analyses will be necessary to decipher the exact role of HRS in targeting BST-2 to the MVB pathway under physiological conditions and after infection by HIV-1 , and the requirement for ubiquitination in these processes . Moreover , it would be interesting to further establish the role of this pathway in Vpu's function in the natural cellular targets of HIV-1 , such as primary CD4 T-cells and macrophages . In summary , we show that HRS facilitates the constitutive turnover of BST-2/tetherin and that Vpu enhances this degradation . HRS is required for Vpu to efficiently down-regulate BST-2 from the surface of the cells and to target BST-2 to the MVB/lysosomal degradation pathway , thereby promoting HIV-1 release . This highlights a role for HRS and the ESCRT/MVB pathway in the regulation of Vpu function . Although the precise molecular and cellular mechanisms by which Vpu connects to the ESCRT machinery to modulate BST-2 expression remains to be elucidated , our data bring significant new insights to understanding how this viral protein uses cellular machineries to counteract host cell restriction and favour HIV-1 dissemination .
HRS and VPS4 cDNA were cloned into pEGFP-C vector ( Clontech , France ) . The VPS4 E223Q mutant was made by PCR mutagenesis using the QuikChange II site directed mutagenesis kit ( Stratagene , France ) . Expression vectors for HRS fused to the Streptavidin binding protein ( SBP ) and Calmodulin binding protein ( CBP ) affinity tags were obtained by cloning HRS into the pNTAP-B vector ( Stratagene , France ) . The expression vector for Flag-tagged BST-2 was obtained by cloning of BST-2 into p3XFlag vector . cDNA for HA tagged-Vpu cloned into pCDNA3 . 1 was a gift from Dr . Florence Margottin-Goguet [45] All mutagenesis and subclonings were verified by DNA sequencing . HeLa , HeLa P4R5 MAGI ( NIH; NIH AIDS Research and Reference Reagent Program , Division of AIDS , NIAID ) , HeLa TZM-bl ( NIH ) and HEK 293T cells were grown in DMEM plus glutamine , antibiotics and 10% decomplemented-FCS ( foetal calf serum ) ( GibcoBRL , Invitrogen ) . HeLa P4R5 MAGI cell cultures were supplemented with 100 µg/mL geneticin and 1 µg/mL puromycin . The 21-nucleotide RNA duplexes designed to target HRS ( 5′ CGACAAGAACCCACACGUCdTdT 3′ at positions 162-180 ) or TSG101 ( 5′ CCUCCAGUCUUCUCUCGUCdTdT 3′ at positions 415-433 ) were previously described [5] , [16] and synthesized by Thermo Scientific Dharmacon ( Perbio Science , France ) . The On-target plus SMART-pool siRNA targeting BST-2 was purchased from Dharmacon ( # L-011817-00 ) . The On-target plus non-targeting siRNA 1 ( # D-001810-01 from Dharmacon ) was used as control . The cells were transfected with 4 to 30 nM siRNA using Lipofectamine RNAiMAX ( Invitrogen ) according to the manufacturer's instructions . Transient transfections of HeLa and HEK 293T cells with mammalian expression vectors were performed using FuGENE ( Roche diagnostics ) following the manufacturer's instructions . Cells were pre-incubated for 45 min with DMEM plus 10 mM Hepes , 1 mg/ml BSA ( Calbiochem , VWR ) and 10 µg/ml Cycloheximide ( Calbiochem ) . The cells were then washed and incubated for the indicated times in regular growth medium plus 10 mM Hepes , 10 µg/ml cycloheximide and 150 ng/ml Epidermal Growth Factor ( EGF ) ( Invitrogen ) where indicated . The cells were harvested , washed in PBS , lysed and analysed by western blotting as described [46] . Stocks of HIV-1 NL4-3 WT ( NIH ) , HIV-1 NL4-3 Udel [47] , VSV-G pseudotyped NL4-3 WT and VSV-G pseudotyped NL4-3 Udel were obtained as described [46] . Viral titres ( Multiplicity of Infection: MOI ) were assessed by infection of the indicator cells HeLa TZM-bl ( bearing the β-galactosidase gene under the control of HIV-1 LTR ) with serial dilutions of the stocks , followed by a β-galactosidase coloration of the cells and counting of blue cells . HeLa P4R5 cells ( 1×105 ) transfected with siRNA were infected with NL4-3 HIV-1 at an MOI of 0 . 005 . Six hours after infection , cells were washed and placed in fresh medium . Supernatants were collected every 24 h for 4 days and used for HIV-1 CAp24 quantification by ELISA . At the end of the experiment , cells were lysed and HRS depletion was assessed by western blotting using rabbit anti-HRS ( Euromedex , France ) and mouse anti-tubulin ( DM1A , Sigma , France ) antibodies . For HIV-1 production assays on a single round of replication , HeLa cells or 293T cells were treated with siRNA as described above . After 48 h , cells were infected with NL4-3 ( WT or Udel ) HIV-1 pseudotyped with VSV-G for 6 hours at an MOI of 0 . 5 . The cells were washed 24 h later , and incubated for a further 24 h . Supernatants were then collected , 0 . 45 µm-filtered , and used for HIV-1 CAp24 quantification by ELISA ( released CAp24 ) . Viral particles released into the supernatant were pelleted through a 20% sucrose cushion by ultracentrifugation at 150 000 g for 90 min , and resuspended in Laemmli buffer . Pelleted viruses were analysed by western blotting using mouse anti-CAp24 ( ARP366 , NIBSC ) . The cell lysates were analyzed by western blotting using mouse anti-CAp24 ( ARP366 , National Institute for Biological Standards and Control ( NIBSC ) , UK ) , mouse anti-MAp17 ( 18A , Hybridolab , France ) , mouse anti-SUgp120 ( 110H , Hybridolab ) , human anti-TMgp41 ( 2F5 , NIH ) , rabbit anti-Vpu ( NIH ) , rabbit anti-Nef ( NIH ) , rabbit anti-HRS ( Euromedex France , Souffelweyersheim , France ) , mouse anti-TSG101 ( BD-Biosciences , France ) , rabbit anti-BST-2 ( NIH ) and mouse anti-tubulin antibodies . HIV-1 CAp24 antigen contained in the cell lysates ( cell-associated CAp24 ) was also quantified by ELISA . In a single round infectivity assay , the titres of released viruses were determined by infection of the indicator cells HeLa TZM-bl in a standardized 96-well titration assay by luminometric analysis of firefly luciferase activity ( Kit luciferase Assay reagent , Promega , France ) following the manufacturer's instructions . The GST-pull down experiment was done using cytosolic extracts of siRNA transfected HeLa cells prepared in 50 mM Tris pH 7 . 6 , 150 mM NaCl , 2 mM EDTA , 0 . 5% ( v/v ) Triton X-100 ( lysis buffer ) . Aliquots ( 2 µg ) of purified GST , GST-p6 or GST-Gag proteins were immobilized on glutathione-Sepharose beads ( GE healthcare ) and incubated for 4 h with 2 mg of cell extract ( diluted in 500 µl of lysis buffer ) . The beads were then washed 3 times in lysis buffer and once in PBS , and bound cellular proteins were separated by SDS-PAGE and revealed by western blotting using antibodies to TSG101 , HRS and tubulin . HEK 293T cells were transfected with ( a SBP-CBP-tagged HRS expression vector ( pNTAP-HRS ) or the corresponding empty vector ( pNTAP ) along with HA-tagged Vpu and Flag-tagged BST-2 expression vectors or with ( b ) SBP-CBP-tagged HRS expression vector along with HA-tagged Vpu and p3X-Flag empty vector or ( c ) Flag-tagged BST-2 expression vector and PCDNA3-1 empty vector . Cells were harvested 24 hours after transfection and lysed following instructions provided by the Interplay mammalian TAP-system ( Stratagene , France ) . The cell lysates were cleared by centrifugation and subjected to two consecutive rounds of pull-down by incubation with a Streptavidin resin , followed by the elution of the bound cellular proteins and incubation of the eluted protein with a Calmodulin resin ( following the manufacturer's instructions ) . The beads were then washed in lysis buffer and the TAP-purified proteins were resolved by SDS-PAGE and revealed by western blotting using antibodies to HRS , HA and Flag . Forty eight hours post-infection , siRNA transfected HeLa cells were harvested by scrapping , washed twice in cold PBS/1% ( w/v ) BSA and stained for 1 hour at 4°C with rabbit anti-BST-2 antibody ( NIH ) or control rabbit IgGs ( rabbit serum; Sigma , France ) in PBS/1%BSA . The cells were then washed three times in PBS/1%BSA , and stained for 1 hour at 4°C with a cy5-conjugated donkey anti-rabbit antibody in PBS/1%BSA . Cells were washed , fixed in 4% paraformaldehyde ( PFA ) , quenched for 10 min in PBS/0 . 1 M glycine , and then permeabilized in PBS/1% BSA/0 . 05% saponin before staining with a FITC-conjugated anti-CAp24 ( KC57-FITC , Beckman Coulter , France ) . Cells were washed and fixed in PBS/1% BSA/1% PFA before analysis using the Cytomics FC500 Flow Cytometer ( Beckman Coulter ) . Gates for FITC were set using non-infected cells . HeLa cells transfected with plasmids encoding the GFP-tagged constructs along with plasmid encoding the provirus NL4-3 wt or Vpu-defective NL4-3 ( NL4-3 Udel ) were harvested 48 hours after transfection and stained for 1 hour at 4°C with anti-BST-2 mouse monoclonal antibody ( Abnova , Tebu-bio; France ) or isotype control mouse IgG1 ( BD-bioscience; France ) , along with human anti-Env antibody ( 2G12 , NIBSC ) , in PBS/1%BSA . The cells were then washed three times in PBS/1% BSA , and stained for 1 hour at 4°C with an Alexa 647-conjugated donkey anti-mouse and PE-conjugated donkey anti-human antibodies in PBS/1% BSA . Cells were washed , fixed in PBS/1% BSA/1% PFA before analysis using the Cytomics FC500 Flow Cytometer . Gates for GFP and PE were set using respectively non-transfected and non-infected cells . All the data were analysed using the CXP cytometer software . Cells , grown on cover slips , were permeabilized in PBS/0 . 1% BSA/0 . 05% saponin [16] before fixation with 4% PFA in PBS for 15 min . PFA-fixed cells were permeabilized and blocked with 0 . 2% BSA/0 . 1% saponin in PBS for 30 min . Cells were incubated for 30 min with mouse anti-BST-2 antibody ( Abnova , France ) alone or together with human anti-Env antibody ( 2G12 , NIH ) in blocking solution , washed and incubated for 30 min with appropriate fluorophore-conjugated secondary antibodies . Cells were analyzed with a Leica SP2 confocal microscope . Series of optical sections were recorded and image processing was performed using Adobe Photoshop CS2 software . Cells were fixed in 4% PFA in 0 . 1 M sodium phosphate buffer , pH 7 . 4 , embedded in gelatine , and frozen for cryosectioning , as described previously [48] , [49] . For immunofluorescence staining , semi-thin ( 0 . 5 µm ) cryosections were placed on glass slides , quenched in 50 mM glycine/50 mM NH4Cl and extracted for 6 min in 0 . 1% Triton X-100 . Sections were stained with antibodies to MAp17 ( 4C9 , ARP 342 , NIBSC ) and Env ( 2G12 , NIBSC ) and goat anti-mouse Alexafluor-488 and goat anti-human-Alexafluor-594 ( Invitrogen ) diluted in PBS/1% BSA , and mounted in Mowiol . Sections were examined with an Axioskop ( Carl Zeiss MicroImaging , Inc . ) ; images were recorded with a CCD camera ( Orca C4742-95; Hamamatsu ) and processed using Adobe Photoshop 8 . For immuno-EM , ultrathin cryosections ( 50 nm ) were stained with mouse antibodies against HIV-1 p24/p55 ( ARP365 and ARP366 , NIBSC ) , a rabbit anti-mouse bridging antibody ( DakoCytomation , Ely , UK ) , and 5 nm PAG ( Protein A gold reagents were obtained from the EM Lab , Utrecht University , Utrecht , The Netherlands ) . Sections were fixed in 1% ( v/v ) glutaraldehyde for 10 min , quenched in 50 mM glycine-50 mM NH4Cl in PBS and incubated with the human anti-Env mAb 2G12 and 10 nm PAG . Sections were embedded in uranyl acetate in methylcellulose , as described previously [49] , and examined with a Technai G2 Spirit transmission electron microscope ( FEI Company UK . Ltd . , Cambridge , UK ) . Digital images were recorded with a Morada 11 MegaPixel TEM camera ( Soft Imaging System ) and the AnalySIS software package . Images were adjusted for brightness and contrast , and figures were assembled with Photoshop 8 . | The release of HIV-1 particles requires a series of interactions between proteins encoded by the virus and key cellular components , including elements of the cellular membrane trafficking apparatus such as the Endosomal Sorting Complexes Required for Transport ( ESCRT ) machinery . This machinery is composed of four multiprotein complexes ( ESCRT-0 , -I , -II and –III ) that are involved in the sorting of ubiquitinylated membrane proteins for lysosomal degradation . Gag , the major structural protein of HIV , recruits the ESCRT-I and III complexes to mediate the scission of budding virions . Following ESCRT-mediated scission of viral particles , the HIV-1 accessory protein Vpu promotes the release of the mature virions by counteracting a cellular restriction factor BST-2/tetherin that physically tethers viral particles to the plasma membrane of infected cells . Here we show that HRS , a component of the ESCRT-0 complex , is required for Vpu to efficiently modulate the expression of BST-2 and promote HIV-1 release , highlighting an additional role of the ESCRT machinery in virus production . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"cell",
"biology/membranes",
"and",
"sorting",
"virology/immunodeficiency",
"viruses",
"virology/virion",
"structure,",
"assembly,",
"and",
"egress"
] | 2011 | The ESCRT-0 Component HRS is Required for HIV-1 Vpu-Mediated BST-2/Tetherin Down-Regulation |
Oct4 is a widely recognized pluripotency factor as it maintains Embryonic Stem ( ES ) cells in a pluripotent state , and , in vivo , prevents the inner cell mass ( ICM ) in murine embryos from differentiating into trophectoderm . However , its function in somatic tissue after this developmental stage is not well characterized . Using a tamoxifen-inducible Cre recombinase and floxed alleles of Oct4 , we investigated the effect of depleting Oct4 in mouse embryos between the pre-streak and headfold stages , ∼E6 . 0–E8 . 0 , when Oct4 is found in dynamic patterns throughout the embryonic compartment of the mouse egg cylinder . We found that depletion of Oct4 ∼E7 . 5 resulted in a severe phenotype , comprised of craniorachischisis , random heart tube orientation , failed turning , defective somitogenesis and posterior truncation . Unlike in ES cells , depletion of the pluripotency factors Sox2 and Oct4 after E7 . 0 does not phenocopy , suggesting that ∼E7 . 5 Oct4 is required within a network that is altered relative to the pluripotency network . Oct4 is not required in extraembryonic tissue for these processes , but is required to maintain cell viability in the embryo and normal proliferation within the primitive streak . Impaired expansion of the primitive streak occurs coincident with Oct4 depletion ∼E7 . 5 and precedes deficient convergent extension which contributes to several aspects of the phenotype .
Oct4 is a homeodomain-containing transcription factor ( TF ) of the POU family required for pluripotency in ES cells and preimplantation embryos [1] . It has been extensively characterized in ES cells , and established as a hub of the signaling network that maintains pluripotency [2]–[5] . Embryonically , Oct4 is present in the developing zygote and down-regulated somatically between E7 . 0 and E9 . 0 depending on the cell type ( see Supplementary ( S ) Figure ( Fig . ) S1 and S2 for detail ) [6] , [7] . After E9 . 0 of murine development Oct4 is restricted to the germline , persisting until maturation of type A to type B spermatogonia in the male germline , in contrast to the female gametic lineage where it is depleted during meiosis ( E14–16 ) before up-regulation as oocytes mature within primordial follicles [6] , [8]–[10] . Several regulators of Oct4 have been established in vivo . Oct4 is maintained through the early stages of embryonic development by intercellular Nodal acting in part through Smad2 [11] , [12] . Conversely , Cdx2 mediates repression of Oct4 in trophectoderm of the early blastocyst , while both Eomes and Gcnf mediate repression in the embryo after implantation [13] , [14] . Oct4 buffers the ICM against differentiation into trophectoderm ( the embryonic contribution to the placenta ) , but the proposal that Pou5f1 ( gene symbol for Oct4 ) emergence relates to evolution of the mammalian placenta [15] is not supported given that Pou5f1 evolved before the origin of amniotes [16] . It is unknown whether Oct4 has a conserved role , or any post-implantation function in murine somatic development . Pluripotent somatic cells persist until E7 . 5–8 . 5 based on teratogenesis experiments [17] , [18] and ∼E8 . 0 based on epiblast stem cell ( EpiSC ) derivation [19] , suggesting that Oct4 might continue to maintain pluripotency during this window of development . in vitro studies have also inferred many roles for Oct4 between the pre-streak and headfold stages , ∼E6 . 0–E8 . 0 , including regulating neural versus mesendoderm differentiation [20] , [21] as well as promoting cardiomyocyte [22] and neuronal differentiation [23] . However aside from maintaining the viability of primordial germ cells ( PGCs ) , Oct4's role in post-implantation development has not been characterized in vivo [1] , [2] , [24] , [25] . The extent of Oct4's function at the molecular level is also unclear . Physical interactions suggest Oct4 may have roles in chromatin modification , regulation of transcription , DNA replication and DNA repair as well as post-transcriptional modification , ubiquitination , and various other functions [2]–[4] , [26] , [27] . Oct4 both activates and represses transcription [28] . It binds thousands of sites in the ES cell genome , often co-occupying these sites with Sox2 , Nanog , Smad1 and Stat3 [5] . The majority of genes occupied by several of these transcription factors ( TFs ) are active in ES cells , but their binding does not ensure expression [5] . Since Oct4 protein normally persists in somatic cells until ∼E7 . 0–E9 . 0 but Pou5f1 null embryos arrest at E3 . 5 , we asked what role Oct4 had later in murine development , using a conditional system to deplete it ∼E7 . 5 . We show that Oct4 depletion ∼E7 . 5 results in craniorachischisis , random heart tube orientation , failed turning , defective somitogenesis as well as posterior truncation . The phenotype is not the result of a general delay in development , nor does it result from a failure in the pluripotency network present in the ICM . Depletion of Sox2 , another core member of the pluripotency network in an overlapping window of development does not phenocopy Oct4 depletion . Instead , Oct4 is required until ∼E7 . 5 to maintain cell viability in the embryo and proliferation in the primitive streak . In its absence , convergent extension is disrupted leading to several morphogenetic defects .
We used a conditional mutant of Oct4 to study its role after E3 . 5 when it is essential for development . We used floxed Pou5f1 alleles ( Oct4f ) [25] and a tamoxifen inducible recombinase ( CreERT2 ) that is ubiquitously expressed from the ROSA locus [29] . To establish the window of development during which embryos are sensitive to Oct4 depletion , we staggered the initial dose of tamoxifen with respect to embryonic maturity and administered a second supplementary dose 12 hrs later to enhance overall recombination efficiency . Oct4f/f;CreERT2+/− embryos administered tamoxifen ∼E8 . 0 and ∼E8 . 5 before analysis ∼E9 . 5 did not have a phenotype ( Table S1 , row A ( S1A ) , while tamoxifen administration ∼E7 . 5 and ∼E8 . 0 before analysis ∼E9 . 5 resulted in a partially penetrant phenotype ( Fig . S3; Table S1B ) . Unlike tamoxifen administration beginning ∼E7 . 5 or ∼E8 . 0 , all Oct4f/f; CreERT2+/− embryos induced ∼E6 . 0 and ∼E6 . 5 before analysis ∼E9 . 5 were amorphous , lacking structures aside from what resembled anterior neural head folds ( Fig . S4; Table S1C ) . Tamoxifen administration ∼E7 . 0 and ∼E7 . 5 also led to a fully penetrant phenotype ∼E9 . 5 ( Table S1D ) . E9 . 5 embryos administered tamoxifen ∼E7 . 0 and ∼E7 . 5 failed to turn , had severe posterior truncations , randomly oriented heart tubes , craniorachischisis ( open neural tube along its entire length ) as well as impaired somitogenesis ( Fig . 1A–C ) . Such animals are referred to as Oct4COND MUT in the remainder of this report . The phenotype is not a consequence of tamoxifen administration , leaky recombinase activity prior to tamoxifen administration , or associated with recombination of a single Pou5f1 allele: no Oct4f/f embryos induced ∼E7 . 0 , no uninduced Oct4f/f;CreERT2+/− embryos , nor any Oct4+/f;CreERT2+/− embryos induced ∼E7 . 0 had phenotypes ∼E9 . 5 ( Table S1E–G ) . Reducing the quantity of tamoxifen per dose administered ∼E7 . 0 or failure to administer the second dose ∼E7 . 5 led to incomplete penetrance of the Oct4COND MUT phenotype ( Table S1H–J ) : 80% , 40% and 0% of embryos ∼E9 . 5 exhibited the Oct4COND MUT phenotype when a single full , half , and quarter tamoxifen dose was administered ∼E7 . 0 ( Table S1H–J ) . This suggests reduced recombination with these lower tamoxifen doses . Collectively , these data support Oct4 depletion causing the Oct4COND MUT phenotype . To determine the time course of Oct4 depletion with this system , we compared Oct4 transcript and protein abundance between Oct4f/f and Oct4f/f;CreERT2+/− littermates administered tamoxifen ∼E7 . 0 . A single dose of tamoxifen was used to avoid a compound effect from a second dose . Relative Oct4 transcript abundance ( Oct4f/f;CreERT2+/−/Oct4f/f;CreERT2−/− littermates ) was significantly different 12 hrs after tamoxifen administration ( ATA ) ( Fig . 1D; Table S1K; F5 , 13 = 15 . 48 , p<0 . 05 1-way ANOVA , *p<0 . 05 , **p<0 . 01 Bonferroni posttest ) . The fraction of cells in which Oct4 was detectable by immunohistochemistry was lower 20 hrs ATA , which is ∼E7 . 5 ( Fig . 1E , Fig . S5A–D; Table S1L; F3 , 10 = 12 , p<0 . 05 1-way ANOVA , **p<0 . 01 Bonferroni posttest ) . A distinct primary antibody indicated that Oct4 protein was undetectable 24 hrs ATA in Oct4f/f; CreERT2+/− embryos ( Fig . 1F , G; Table S1L ) . Since penetrance of the phenotype is complete when tamoxifen administration begins ∼E7 . 0 , partial when tamoxifen administration begins ∼E7 . 5 , and the fraction of cells with detectable Oct4 protein reduced ∼20 hrs ATA ( following administration ∼E7 . 0 ) , these data indicate that Oct4 is required until ∼E7 . 5 . Oct4 depletion does not cause a global delay in development . Administering tamoxifen ∼E7 . 0 and ∼E7 . 5 to avoid partial penetrance , Oct4f/f;CreERT2+/− embryos were recovered in a ratio of 1∶1 with Oct4f/f littermates until E9 . 5 , but less frequently at E11 . 5 ( Fig . 2A; Table S1M–O ) . Features disrupted in Oct4COND MUT remained arrested in the mutants that persisted beyond E9 . 5 ( Fig . 2B , C ) , indicating that the Oct4COND MUT phenotype is not a global delay in development but disruption of select features . Indentation of the otic cup occurred and the branchial arches formed in Oct4COND MUT , events that normally occur by E9 . 0 . Forelimb buds also protruded in Oct4COND MUT as they normally do by E9 . 5 . Conversely , the neural tube normally closes rostrally between E8–9 and caudally by E9–10 ( we refer to caudal and rostral neural tube closure with respect to closure point 1 at the hindbrain cervical boundary throughout; see Figure 2D ) [30] , turning normally occurs by ∼9 . 0 and posterior extension normally reaches 21–29 somites by E9 . 5 in WT embryos . These events always failed at E9 . 5 when Pou5f1 excision was induced ∼E7 . 0 ( Fig . 1A–C; Table S1D; 26 . 5 versus 4 . 6 somites in Oct4f/f versus Oct4f/f;CreERT2+/− littermates ) . Additionally , heart tube orientation was randomized , 38 . 6% of Oct4f/f;CreERT2+/− had situs inversus while the orientation of 6 . 8% was ambiguous ( Table S1P; p>0 . 05 Chi-square test ) . The neuroepithelium of Oct4COND MUT embryos was also thicker in regions , particularly in the distal portion of the embryo ( Fig . S6A–C; Table S1D; F1 , 287 = 94 . 95 , p<0 . 05 2-way ANOVA , ***p<0 . 001 Bonferroni posttest ) . These data indicate that Oct4 is required for posterior extension , turning , heart tube orientation and neural tube closure ( NTC ) . Partial phenotype penetrance following tamoxifen administration ∼E7 . 5 was used to assess whether the cause of disrupted features in Oct4COND MUT embryos were related . Coincidence of features in litters with incomplete phenotype penetrance suggests related causation of the coincident features . Craniorachischisis and posterior truncation coincided in all 23 of the 36 embryos analyzed ( Fig . S2; Table S1B; p = 1 . 64E-10 , hypergeometric test ) . Conversely 2 turning defects in the 9 embryos where rostral NTC failed suggests independence of these processes , although the small number of embryos limits statistical power in this case ( Fig . S3; Table S1B; p = 0 . 72 , hypergeometric test ) . These data suggest independent requirements for Oct4 in closure at closure point 1/posterior extension and rostral NTC . Craniorachischisis occurs when closure at closure point 1 fails ( see Figure 2D ) . Convergent extension elongates the embryo in the anterior-posterior axis during gastrulation and neurulation , bringing the neural folds into opposition prior to adhesion at closure point 1 . Failed convergent extension results in broad midlines and enlarged notochord diameter as both narrow during convergent extension . Oct4COND MUT embryos exhibit broad neural plates ( Fig . 2H–J; Table S1D; F2 , 22 = 17 . 42 , p<0 . 05 2-way ANOVA , **p<0 . 01 Bonferroni posttest ) and enlarged notochord diameter ( Fig . S6D–F; Table S1D; p<0 . 05 , two-tailed student t-test ) . Concordance between posterior truncation and craniorachischisis , broadened neural plates , and broader notochords are consistent with deficient convergent extension . NTC rostral and caudal to closure point 1 occur by different mechanisms . Unlike the spinal region where expansion of paraxial mesoderm is not required for elevation and subsequent NTC , cranial NTC is initiated by expansion of underlying mesenchyme [30] . Mesenchyme density , including cranial mesenchyme , was reduced in Oct4COND MUT ( Fig . 2E–G; Table S1D; F1 , 13 = 54 . 59 , p<0 . 05 2-way ANOVA , *p<0 . 05 , ***p<0 . 001 Bonferroni posttest ) . Hence expansion of cranial mesenchyme that is required for cranial NTC is deficient in the absence of Oct4 . A requirement for Oct4 in extraembryonic tissue offers one possible explanation for the Oct4COND MUT phenotype: ∼E7 . 5 Oct4 is present in extraembryonic mesoderm , allantoic angioblasts as well as extraembryonic endoderm which promotes proliferation and organization of the primitive streak [6] , [31] . To test this possibility , Oct4+/+ Red fluorescent protein positive ( RFP+ ) ES cells were aggregated with tetraploid Oct4f/f;Z/EG+/−;CreERT2+/− embryos , where ES cells contribute to the embryo , and tetraploid cells generate trophectoderm and visceral endoderm [32] . In this scheme , tamoxifen administration will selectively remove of Oct4 from the tetraploid extraembryonic lineages . Tetraploid Oct4f/f;Z/EG+/−;CreERT2+/− embryos induced ∼E6 . 5 and ∼E7 . 0 supported development of WT ES-derived embryos to E9 . 5 ( Fig . 3A–C , E; Table S1Q ) . Embryos were dosed on this relatively early schedule to avoid false negatives that might result from altered timing of development associated with transferring embryos to pseudopregnant mothers . In practice transferred embryos synchronize with the maternal uterine environment [33] , suggesting false negatives for this reason are unlikely . Normal embryonic development after excision of Pou5f1 in trophectoderm and visceral endoderm suggests Oct4 is required in embryonic tissue . To identify non-autonomous effects of Oct4 depletion , we tested whether lineage-specific removal of Oct4 affected development of other tissues . Since Oct4 is present in the primitive streak , neuroepithelium and portions of mesoderm ∼E7 . 5 as well as mosaically in definitive endoderm ( Fig . S1 and S2 ) , a primary effect in one of these lineages might non-autonomously cause other aspects of the Oct4COND MUT phenotype [6] . To test this possibility , Oct4 was removed in the neuroepithelium using Sox1-Cre , which is expressed and catalytically active from ∼E7 . 5 [34]; in definitive endoderm using tamoxifen-inducible Foxa2mcm , which is expressed ∼E6 . 25 [35]; as well as in embryonic mesoderm using Brachyury ( Bry ) -Cre , which is expressed and catalytically active from ∼E6 . 25 [36] . Excision of Pou5f1 by lineage-specific recombinases ( Bry-Cre , Sox1-Cre or Foxa2mcm ) did not result in a phenotype or impact embryonic viability at E9 . 5 . Oct4f/f; Z/EG+/; lineage-specific Cre+/− embryos should reveal aspects of the Oct4COND MUT phenotype related to requirements for Oct4 within their respective expression domains or cause the embryo to resorb by E9 . 5 if development is more severely impacted than in Oct4COND MUT embryos . Recombination at the lacZ/enhanced GFP ( Z/EG ) locus yields GFP expression , so the Z/EG allele was incorporated to gauge recombination efficiency [37] . Based on the parental genotypes used in the cross ( Table S1R–T ) , a genotypic ratio where Oct4f/f; Z/EG+/−; lineage-specific Cre+/− embryos comprise ¼ of the progeny is expected if this genotype , where lineage-specific excision of Pou5f1 occurs , does not impact viability . Such embryos with no phenotype comprised ¼ of each litter ( Table S1R–T ) . To test whether the lineage-specific recombinases yielded false negative results due to infrequent biallelic excision , we assessed the development of embryos where one Pou5f1 allele was removed prior to recombinase expression . Even with this sensitized approach , Oct4Δ/f; Z/EG+/−; lineage-specific Cre+/− embryos with no phenotype comprised ¼ of the progeny at E9 . 5 . This genotypic ratio indicates that excision of Pou5f1 by these lineage-specific recombinases did not impact viability ( Table S1U , V ) . Since false-negatives may arise due to low recombination efficiency in this scheme , we used the GFP expression resulting from recombination at the Z/EG locus in Oct4f/+; Z/EG+/−; lineage-specific Cre+/− embryos as a proxy for recombination efficiency . By E9 . 0 Sox1-Cre and Bry1-Cre induced >95% and >51% recombination within their respective domains ( Fig . S7A–C; Table S1W–Y ) , while Foxa2mcm yielded <5% ( data not shown ) . However , prior to E8 . 0 when embryos are sensitive to Oct4 depletion , Sox1-Cre and Bry-Cre also yielded <5% recombination ( Fig . S7C; Table S1Z , AA ) [30] . Notably , the distribution of Oct4Δ/f; Z/EG+/−; Bry-Cre+/− cells did not appear altered ∼E9 . 5 ( Fig . S7D , E ) , suggesting that any effect Oct4 has on cell fate either coincides with lineage specification or precedes it . To investigate how recombination frequency influences phenotype penetrance in embryos where Pou5f1 is removed by lineage-specific recombinases , we generated diploid chimeras by aggregating WT and Oct4f/f;HisGFP+/−;CreERT2+/− morulas . The ubiquitously expressed fusion protein ‘HisGFP , ’ which is comprised of histone H2B and eGFP was used to mark transgenic cells [38] . Following tamoxifen administration ∼E6 . 5 and ∼E7 . 0 , we recovered 16 chimeras where contribution by Oct4f/f;HisGFP+/−;CreERT2+/− morulas ranged from 20–60% ( Table S1AB ) . 11 of these 16 embryos had no phenotype , while the remaining 5 chimeras had rostral NTC deficits ( Fig . 3D , E ) . This indicates that Oct4+/+ cells rescue the developmental deficiencies caused by Oct4−/− cells in mosaic embryos . Since efficient depletion of Oct4 is required for the Oct4COND MUT phenotype , the inefficient recombination of Bry-Cre , Sox1-Cre and Foxa2mcm during the window of development in which embryos are sensitive to Oct4 depletion does not resolve whether Oct4 is ubiquitously required ∼E7 . 5 , required only in unspecified progenitors , or necessary in a subset of specified lineages , such as in specified Oct4+Bry+ mesoderm . Since this data suggested that differences in the kinetics of Pou5f1 excision with lineage-specific recombinases and CreERT2 ( when tamoxifen is administered ∼E7 . 0 ) are responsible for the absence and presence of phenotypes following Pou5f1 excision , we tested whether expansion of specified lineages was affected in Oct4COND MUT embryos . Lineage-specified Bry+ and Sox2+ cells were present 48 hrs ATA in Oct4f/f;CreERT2+/− embryos ( Fig . S8A , B; Table S1AC ) . We quantified the fraction of phosphorylated Histone H3 ( PH3 ) + cells in specified lineages . The PH3+ fraction of neural or mesoderm cells ( Oct4f/f;CreERT2+/− versus Oct4f/f ) was the same ( Fig . S8C , Table S1AC ) . The data indicate that expansion of these specified lineages is not impacted by Oct4 depletion . To test whether disruption of the pluripotency network causes the Oct4COND MUT phenotype , we removed Sox2 using the same conditional approach [39] . Sox2 is a core component of the pluripotency network that complexes with Oct4 , co-occupies many genomic sites ( Oct4/Sox2 ) and is required for maintenance of Pou5f1 expression in ES cells . ES cells differentiate into trophectoderm when Sox2 is removed [40] , however the ability of Oct4 over-expression to rescue pluripotency in these cells suggests that the critical role of Sox2 in pluripotency is to maintain Pou5f1 expression [40] . Sox2 null embryos lack epithelial cells typical of the epiblast and have a later extraembryonic defect which does not permit development past E7 . 5 [41] . Following tamoxifen administration ∼E6 . 5 and ∼E7 . 0 to Sox2f/f;CreERT2+/− embryos [39] , hydrocephalus was evident in 11/20 Sox2f/f;CreERT2+/− and 2/20 others had kinked neural tubes ∼E9 . 5 ( Fig . 4A–C; Table S1AD ) . Thus Sox2 removal did not phenocopy Oct4 depletion ∼E7 . 5 . These data do not rule out partial compensation for loss of Sox2 by redundant factors , however between E7 . 0–E8 . 0 Oct4 and Sox2 only overlap spatially in anterior neuroepithelium ( compare Figure S1 , S2 and S9 ) [6] , [41] . The distinct phenotypes produced by depletion of Sox2 and Pou5f1 indicate that at least part of their functions do not overlap ∼E7 . 0–E8 . 0 , in contrast to ES cells . Oct4 is reported to bind 784–4234 genomic loci in ES cells depending on the methodology used to map binding sites [5] , [42] , [43] . To determine which targets might be contributing to the Oct4COND MUT phenotype , we measured gene expression changes that occurred coincident with Oct4 depletion ( ∼E7 . 5 ) and thereafter ( ∼E8 . 0 and ∼E8 . 5 ) . Oct4f/f;CreERT2+/− embryos were separated from Oct4f/f littermates by genotyping extraembryonic tissue , and differential expression assessed within litters with ≥3 CreERT2+/− and ≥3 CreERT2−/− embryos ( Table S1AE ) . RNA was extracted 24 , 36 and 48 hrs ATA , when Oct4 transcript abundance in CreERT2+/− embryos is <5% CreERT2−/− littermates ( Fig . S5A–D ) . 754 unique genes were differentially expressed ( p<0 . 01 ) at one or more of these three timepoints . To determine whether the differential expression following Oct4 depletion was a direct consequence of Oct4 loss at its genomic targets , we assessed whether Oct4's direct targets were enriched amongst up- or down-regulated genes as Oct4 both activates and represses transcription [28] . Systematic mapping of TF targets in early embryos is currently prohibitive [44] , so a genome-wide binding map of Oct4 in ES cells was used [5] . This particular genomic binding map , which is based on ChIP-seq data , was used because it offers more complete genomic coverage than target maps based on ChIP-chip data , and also contained the most extensive set of other TF binding maps for additional analysis ( alternatives include: [42] , [43] ) . Enrichment of TF binding targets from ES cells amongst differentially expressed genes after ∼E7 . 5 requires that binding sites be conserved between these stages . Oct4 binding sites from ES cells were enriched amongst up-regulated genes ( Fig . 5B ) , supporting conservation of the binding sites between ES and ∼E7 . 5–E8 . 5 embryos . Oct4 binding targets were also enriched when alternative datasets were analyzed . For comparison , with the aggregate of differentially expressed genes ( 24 , 36 and 48 hrs ATA ) , enrichment using hypergeometric tests were: p = 3 . 45E-11 [43] , p = 2 . 13E-08 [5] , and p = 7 . 36E-4 [42] . This suggest that the expression changes at these sites were a direct consequence of Oct4-mediated transcriptional regulation being removed after ∼E7 . 5 . Oct4 targets whose transcription is regulated by Oct4 in ES cells were differentially expressed coincident with Oct4 depletion ∼E7 . 5 . Lefty1 and Klf2 that are activated by Oct4 in ES cells decreased [45] , [46] , while Xist was notable among the most up-regulated genes following Oct4 depletion as it is repressed by Oct4 in ES cells [41] . An unbalanced male∶female ratio in the intra-litter comparisons , rather than Oct4 depletion , might explain the increase in Xist transcript abundance since embryos were not sexed in the microarray , however Quantitative ( Q ) -PCR on independent balanced comparisons confirmed that the increase related to Oct4 depletion . An intra-litter comparisons to match developmental stage , and inter-litter comparisons to reduce biological variance associated with comparing a small number of embryos both supported Oct4-mediated repression of Xist ∼E7 . 5: Xist was 3 . 20 times more abundant in the intra-litter comparison , and 2 . 85±0 . 76 s . e . m . more abundant in the inter-litter comparison of Oct4f/f;CreERT2+/−/Oct4f/f 24 hrs ATA ( Table S1AF ) . Enrichment for genomic targets of Oct4 is expected with this approach , but transcriptional activators of Oct4 and proteins that physically interact with it were also differentially expressed . Ligands that maintain Oct4 such as Nodal and Wnt3a [11] , [47] exhibit decreased transcript abundance coincident with Oct4 depletion ∼E7 . 5 , while transcriptional activators of Oct4 such as Sp1 [48] and Ago2 [49] exhibited increased transcript abundance , perhaps due to a feedback loop . Proteins that physically interact with Oct4 were also enriched amongst the genes up-regulated following Oct4 depletion ( see Table S2 for cofactor identities; p = 1 . 99E-08 24 hrs ATA , p = 1 . 64E-05 36 hrs ATA , p = 5 . 55E-07 48 hrs ATA enrichment using hypergeometric tests ) . Interestingly , we found considerable enrichment for Oct4 within genomic regulatory elements of these physical cofactors ( p = 5 . 34E-07 for 24 , 36 and 48 hrs ATA collectively using a hypergeometric test ) . This suggests that ∼E7 . 5 Oct4 directly represses expression of a subset of the genes it physically interacts with in ES cells and that its absence triggers positive indirect feedback of the expression of others . Collectively , these data suggest that several regulatory relationships of Oct4 are maintained between preimplantation development and ∼E7 . 5–8 . 5 . To test whether signaling networks other than direct targets of Oct4 might contribute to the Oct4COND MUT phenotype , we determined the transcriptional response that target sets bound by TFs other than Oct4 had to Oct4 depletion . The binding maps of 12 other TFs , and combination of several with Oct4 , were assessed for enrichment amongst the genes differentially expressed after Oct4 depletion ( Fig . 5A ) [5] . Targets of c-Myc and Smad1 were enriched amongst genes up-regulated after Oct4 depletion [5] . Unlike c-Myc , which does not cluster at binding sites with Oct4 in the genome , Oct4 facilitates the binding of Smad1 such that they overlap at a subset of sites [5] . However up-regulation of Smad1 targets after Oct4 depletion occurred at sites Smad1 occupies independent of Oct4 , indicating that enrichment of up-regulated Smad1 targets is not due to direct relief of Oct4-mediated repression at sites that the two co-occupy [5] . The enrichment of Smad1 targets amongst up-regulated genes that are not co-occupied by Oct4 are: p = 6 . 14E-06 24 hr ATA , p = 4 . 55E-03 36 hr ATA , p = 3 . 53E-09 48 hr ATA ( hypergeometric test ) . Like Oct4 , Smad1 has been implicated in both activation and repression of target genes [50] , consistent with a separate subset of Smad1 targets are de-repressed 24 hrs ATA . These data suggest that the absence of Oct4 yields a transcriptional environment conducive to target activation by c-Myc and Smad1 . Conversely , enrichment of co-occupied Oct4/Sox2 target sites amongst down-regulated genes ( Fig . 5C ) suggests that Oct4 participates in transcriptional activation of these ∼E7 . 5 and after . Since conditional removal of Sox2 and Pou5f1 do not phenocopy ( compare Figure 1A to 4A ) , Sox2 is either not essential for activation of these sites , which is consistent with data from ES cells [40] , or down-regulation of these targets does not contribute to the Oct4COND MUT phenotype . Oct4 binds thousands of sites in the genome , and it is unlikely that disruption of a single target gene causes the Oct4COND MUT phenotype . To relate molecular changes resulting from Oct4 depletion with the Oct4COND MUT phenotype , we determined which signaling pathways were disrupted coincident with Oct4 depletion and prior to the onset of the phenotype . Unsupervised clustering was used to assess the function of differentially expressed genes collectively . To discern primary effects of Oct4 depletion , we sub-setted for genes that are direct targets of Oct4 based on the ES binding maps [5] , clustered these ( Fig . 6A; Table S1AE ) , and then compared the clusters to global changes ( Fig . 6B; Table S1AE ) . 3 of the 4 pathways showing the strongest enrichment in the set of direct targets also showed significant enrichment in the global set . Coordinate regulation of additional genes that are not targets of Oct4 within the same pathways as those directly regulated by Oct4 , suggests amplification of the direct effects ( Fig . 6A , B; Table S1AE ) . QPCR on independent biological samples confirmed a subset of changes from the global expression analysis ( Fig . 6C , Table S1AG ) , supporting the reproducibility of the differential expression . Differential expression was then considered in relation to the Oct4COND MUT phenotype . The expression profiling suggested that decreased TGF-β signaling and increased nuclear import of NF-κB were primary effects as they occurred within hours of Oct4 depletion ( 24 hrs ATA ) amongst direct targets of Oct4 , while decreased Notch signaling and increased protein translation are other candidates that occurred later ( Fig . 5A ) . The node is required to coordinate left-right asymmetry , specification of definitive endoderm and somitogenesis [51] . Given these roles in development , we considered the possibility that Oct4 was required in node formation a candidate that might explain the situs inversus , defective somitogenesis and the posterior truncation ( via either endoderm specification or defective somitogenesis ) observed in Oct4COND MUT embryos . Gene expression changes following Oct4 depletion also suggested the possibility of node malformation: decreased Dll1 contributed to the ‘Notch signaling’ enrichment in the microarray and was confirmed by QPCR in separate litters ( Fig . 6C; Table S1AG ) . Decreased Dll1 following Oct4 depletion is relevant because loss of Dll1 was previously shown to disrupt node formation and cause defects in left/right asymmetry [52] . While these data were suggestive of a candidate mechanism underlying the Oct4COND MUT phenotype , the presence and appropriate localization of the node marker Chordin both 24 hrs ATA ( Fig . 7A , B , Table S1AC ) and 36 hrs ATA ( Fig . S10A , B; Table S1AC ) suggests that initial node specification occurs in Oct4COND MUT [53] . The disruption of left-right asymmetry is likely downstream of node specification , as transcript abundance of laterality specifiers that are asymmetrically distributed by the node during development is altered: Nodal , Dll1 , Lefty1 and Lefty2 are decreased while Hand1 and Hand2 are increased . These data do not support the Oct4COND MUT phenotype being caused by a failure in Notch-mediated node specification . Contraction of actin-myosin microfilaments contributes to the morphogenetic processes of turning and convergent extension . A decrease in ‘actin filaments’ ( p = 1 . 88E-07 ) following Oct4 depletion ( Fig . 6B; Table S1AE ) suggests that actin networks are affected by Oct4 depletion . The distribution of actin appeared altered 24 hrs ATA with phalloidin staining ( Fig . S10C , D; Table S1AC ) . Indeed the distribution of actin in Oct4f/f;CreERT2+/− embryos suggests that adhesion between anterior and posterior neuroepithelium in the distal portion of the embryo may contribute to thicker neuroepithelium in this regions and impaired embryonic morphogenesis . TGF-β signaling has also been implicated in several processes disrupted in Oct4COND MUT embryos: expansion of primitive streak [54] , patterning derivatives of the anterior primitive streak [55] , establishment of definitive endoderm [56] , maturation of the node [57] and left/right asymmetry establishment [58] , [59] . Unsupervised clustering indicates that Oct4 directly maintains TGF-β signaling ( Fig . 6A ) . TGF-β signaling through Smad2 competes with Smad1 for the co-activator Smad4 [60] , so up-regulation of Smad1 targets following Oct4 depletion may involve an increase in Smad1 , expansion of the domain of activated phosphorylated-Smad1 ( p-Smad1 ) , or diminished competition from TGF-β-Smad2 . Increased transcript abundance of Smad1 was confirmed by Q-PCR ( Fig . 5C; Table S1AG ) . The p-Smad1 domain also appears altered 24 hrs ATA ( Fig . 7C , D; Table S1AC ) . Variance in p-Smad1 introduced by differences in embryonic stage and ‘batch effects’ during detection prohibited making a statistically meaningful quantitative comparison of protein abundance between stage-matched Oct4f/f; CreERT2+/− and Oct4f/f embryos . Quantitative comparison with high-content image analysis software did suggest a difference in p-Smad1 abundance related to Oct4 depletion ( Fig . S11 ) , but this approach would require a considerable increase in sample size to test significance . These data suggest a direct effect of Oct4 depletion on diminished TGF-β signaling . Presence of Oct4 in the primitive streak ∼E7 . 5 ( Fig . S1 ) , impaired axial extension in Oct4COND MUT embryos and differential expression of TGF-β signaling that is essential for expansion of primitive streak [54] suggested an effect on its expansion . An effect on the primitive streak and consequently its derivatives might have broad relevance: cranial mesenchyme supports NTC , while mesendoderm facilitates posterior extension , somitogenesis and turning . The frequency of cells undergoing apoptosis ( Caspase-3+ ) in the Oct4COND MUT was increased ( Fig . 7I; Table S1AC ) , suggesting that diminished cell viability might contribute to the phenotype . Notably , the distribution of apoptotic cells throughout the embryo , including regions where Oct4 is not expressed , suggests that some apoptosis may be a secondary defect . Conversely , fewer cells proliferated indicated by phosphorylated histone H3 positive ( PH3+ ) in the primitive streak of embryos 24 hrs ATA ( Fig . 7G , H , J; Table S1AC ) . To confirm the localization of these effects , we divided embryos into three segments ( proximal anterior , distal and proximal posterior ) and quantified the abundance of transcripts regulating apoptosis and proliferation . To obtain sufficient material for comparison , CreERT2+/−;Oct4f/f samples 24 hrs ATA were compared to CreERT2+/−;Oct4f/f stage-matched samples from separate litters . While there was no difference in the transcript abundance of apoptosis regulators Bax and Bcl2 , a negative regulator of proliferation , Cdkn1c , which exhibited increased transcript abundance in the differential expression analysis was selectively increased in the posterior third of embryos coincident with the loss of Oct4 ( Fig . 7K; Table S1AH ) . These data indicate that ubiquitous Oct4 depletion leads to increased apoptosis and deficient proliferation in the primitive streak .
∼E7 . 5 , Oct4 is still present in the primitive streak , posterior visceral endoderm , several mesoderm derivatives , neuroepithelium as well as extraembryonic endoderm and mesoderm ( Fig . S1 ) [6] . Proliferation of the primitive streak decreases and apoptosis increases within the embryo coincident with Oct4 depletion ∼E7 . 5 , and by ∼E9 . 5 several morphogenetic processes are disrupted: turning , posterior extension , laterality and NTC all are affected , demonstrating that Oct4 is required for somatic development after implantation . Reduced proliferation in the primitive streak coincident with Oct4 depletion suggests that Oct4 might maintain potency ∼E7 . 5 as it does in the ICM [1] . EpiSC-derivation and teratoma assays support the persistence of pluripotent somatic cells ∼E8 . 0 , while lineage tracing indicates the presence of neuro-mesodermal progenitors ∼E8 . 0 [61] . However excision of pluripotency factors Sox2 and Oct4 ∼E7 . 0 do not phenocopy as their depletion in ES cells do [1] , [40] , indicating that the pluripotency network is altered between the ICM and ∼E7 . 5 . Differences in localization contribute: at the latest stage embryos are sensitive to Oct4 depletion and a proliferation deficit is evident in the primitive streak of Oct4COND MUT embryos ( ∼E7 . 5 ) , Sox2 transcript is limited to the chorion and anterior neuroectoderm ( Fig . S9 ) [41] . Neural-specific Sox2 excision results in enlarged lateral ventricles ∼E19 . 5 due to decreased proliferation of neural stem and progenitor cells [62] , suggesting that hydrocephalus in Sox2COND MUT embryos may result from insufficient expansion/thickening of the neuroepithelium . This might render the neuroepithelium more elastic and distended as a result of the positive fluid pressure in the neural lumen [63] , or precede the collapse or kinking of neural tubes that infrequently occurred . The differing phenotypes following depletion ∼E7 . 5 indicate that Sox2 is not required for Pou5f1 transcription or as a cofactor in the processes disrupted in Oct4COND MUT embryos . Oct4 promotes mesoderm as opposed to neural fate during ES differentiation [20] , as does XlPou91 ( the paralog in X . laevis ) in response to FGF [64] , [65] , suggesting that Oct4 depletion might divert mesoderm to neural tissue . Decreased expression of Tbx6 [66] and Wnt3a [67] whose loss is associated with diversion to ectopic neural tubes from paraxial mesoderm following Oct4 depletion is consistent with this possibility , as is thicker neuroepithelium of Oct4COND MUT embryos near closure point 1 . However this differential expression may not reflect altered specification per se , but altered proportions of the embryo associated with defective axial extension . Similarly , neuroepithelial thickening unrelated to cell fate divergence is common amongst mutants with NTC defects such that this is not a reliable indicator of fate changes [30] . Finally , the distribution of Oct4Δ/f; Z/EG+/−; Bry-Cre+/− cells did not appear altered . This suggests that any effect Oct4 has on cell fate either coincides with lineage specification or precedes it . An alternative to an effect on cell fate specification is that Oct4 promotes expansion of unspecified progenitors by driving the cell cycle . Reduced mesenchyme density , decreased proliferation in the primitive streak , increased Trp53 ( p53 ) expression and increased Cdkn1c expression in the Oct4COND MUT embryonic posterior all indicate that expansion of posterior progenitors is disrupted when Oct4 is depleted . The G1/S transition is effectively absent from ES cells , and binding of Oct4 to micro-RNAs that suppress inhibitors of the G1/S transition [68] may promote its bypass and limit the window for lineage-specific chromatin remodeling . Indeed , genes regulating ‘chromatin modification’ are up-regulated 24 hrs ATA coincident with reduced proliferation in the primitive streak ( cluster 1–295: p = 2 . 1E-04 and cluster 613–908: p = 3 . 7E-04 using hypergeometric tests ) . Finally , c-Myc activates G1/S checkpoint complexes [69] , [70] , suggesting that c-Myc may be required to promote G1/S transition when the G1/S checkpoint is established coincident with Oct4 depletion . Morrison and Brickman proposed that the evolutionarily conserved role of Oct4 might be facilitating expansion of progenitor populations during and after gastrulation based on work with paralogs: Pou2 in D . rerio and XlPou91 in X . laevis [64] . These D . rerio Pou2 mutants [71] and X . laevis embryos treated with morpholinos against XlPou91 share posterior truncations [64] . Since Pou5f1 arose by duplication of Pou2 [64] , these data support a conserved role for Oct4 in posterior extension , which in mice includes maintaining proliferation in the primitive streak .
All procedures were approved by the University of Toronto Animal Care Committee in accordance with the Canadian Council on Animal Care . Foremost , both euthanasia and surgery were minimized . When performed , stress was minimized to the greatest extent possible before rapid depressive action on the CNS during euthanasia . Minimally invasive surgeries were performed under anesthetic to achieve complete depression of feedback from the PNS and analgesic used for recovery . For staging , embryos were assumed to be 0 . 5 days post coitum at 1pm on the day a vaginal plug was found . This is 12 hrs after the midpoint of the 14 hr light/10 hr dark cycle we used , where the lights were shut off every night at 8 pm and came on every morning at 6 am . Given the relevance of staging to this set of experiments , it is important to note that use of vaginal plugs –as opposed to direct observation of conception– is accompanied by ±7 hrs of variability in embryonic staging and is inferred from the midpoint of the dark period in the light/dark cycle . Embryos were dissected in Dulbecco's PBS ( Gibco ) and immediately placed in either liquid nitrogen ( for microarrays and QPCR analysis ) or in 4% paraformaldehyde ( for sectioning and immunohistochemistry ) . Dissections for embryonic stages that are whole numbers ( e . g . E8 . 0 or E9 . 0 ) were performed between 9 and 11 pm , while those occurring 12 hrs apart from whole days post coitum ( e . g . E9 . 5 or E10 . 5 ) were performed between 12 and 2 pm . For the experiments assessing the timeframe of Oct4 depletion ( Fig . 1D–G , S5A–D ) , tamoxifen was administered at 9 pm±30 min , and dissections performed the indicated number of hours ATA , e . g . dissections for the time-point 3 hrs ATA were done at midnight ( 12 am ) . The following stocks were used in the study: CD1 ( Charles River ) , Oct4f/f [25] , lacZ/eGFP ( Z/EG ) [37] , B6 . Cg-Tg ( Hist1H2BB/Egfp ) 1Pa/J ( Histone H2B/eGFP fusion ‘HisGFP’ ) [38] , Bry-Cre [36] , Sox1-Cre [34] , Foxa2tm2 . 1 ( cre/Esr1* ) Moon/J [35] , Sox2f [39] , CreERT2 [29] . Individual embryos or the associated extraembryonic tissues were genotyped as originally described . Because a variety of experimental permutations were used in this project , the details of each permutation , including the mouse strains , genotypic ratios , tamoxifen administration regimen and other relevant features are provided on a separate row in Table S1 ( the relevant row is noted as the experiment is described where ‘S1 , row A’ is ‘S1A’ ) . Tamoxifen was administered according to the protocol optimized following CreERT2 development [29] . 99 mg of tamoxifen ( Sigma ) was dissolved by sonication in a solution of 100 ul of ethanol ( Sigma ) and 1 ml of peanut seed oil ( Sigma ) [29] . The solution was kept in a ∼50°C water bath during preparation and prior to administration to avoid precipitation . 50 µl doses of this solution were administered to pregnant mothers by oral gavage using a 250 µl gastight #1725 syringe ( Hamilton ) [29] . Because of the uncertainty associated with staging embryos with vaginal plugs ( ±7 hrs ) , the time-point ( s ) indicated for tamoxifen administration are approximations , and listed as such ( ∼ ) within the text to reflect this uncertainty . In practice , tamoxifen was given at 9pm±30 min ( ∼E6 . 0 , ∼E7 . 0 or ∼E8 . 0 ) or 9 am±30 min ( ∼E6 . 5 , ∼E7 . 5 or ∼E8 . 5 ) . The time-point ( s ) when tamoxifen was administered for each experimental permutation are listed in Table S1 as well as in the figure captions . The density of mesenchyme , frequency of apoptosis and proliferation , relative abundance of transcripts ( other than Oct4 ) , distance between neural folds and thickness of neuroepithelium were compared using 2-way ANOVAs . Depletion of Oct4 protein and transcript were compared with 1-way ANOVAs . F-values from the embryonic genotype's contribution ( Oct4f/f versus Oct4f/f;CreERT2+/− ) to variation are indicated except for Figure 7K and S6C where the intra-embryo segment contribution is reported ( e . g . difference between segments in the same embryo ) . Binding enrichment amongst differentially expressed genes and common causality of disrupted features in partially penetrant Oct4COND MUT embryos was assessed using hypergeometric tests . The thickness of notochords was compared using a two-tailed t-test . A threshold of p<0 . 05 was used for each test ( ANOVA , hypergeometric and t-test ) . Please see the Supplementary Methods ( ‘Text S1 , page 1’ ) for detail on how measurements of Oct4 protein depletion , mesenchyme density , neuroepithelium thickness , notochord thickness , distance between the neural folds , and the fraction of Ph3+ , Caspase-3+ and Oct4+ cells were taken ( ‘Basic Measurements’ ) . Images in Figure 1F , G; Figure 2 F , G , I , J; Figure 3A–D; Fig 7A–J; Figure S6A , B , D , E; Figure S7A , B , D , E; Figure S8A , B and Figure S10A–F were taken with a Zeiss Axio Observer , images of Figure S5A–D were taken with an Olympus Fluoview 1000 , images of Figure 2 B , C and Figure 4A–C were taken with an Olympus SZ61 , and images of Figure 1A–C; Figure 3SA , B and Figure S4A were taken with a Leica MZ16 FA stereomicroscope . Contrast of the images in Figure 3D , 4A and 4C was enhanced with Adobe Photoshop v12 . Oct4 staining was performed as described previously [6] . For all other immunohistochemistry , embryos were fixed in 4% PFA overnight at 4°C , sectioned at a thickness of 10 µm and primary antibodies applied overnight at 4°C at the following concentrations: Oct-3/4 1∶200 ( C-10 Santa Cruz ) , Chordin 1∶100 ( R & D Systems ) , p-Smad1 1∶400 ( Cell Signaling ) , Caspase-3 1∶500 ( Promega ) , Ph3 1∶500 ( Cell Signaling ) , Bry 1∶50 ( R & D Systems ) , Sox2 1∶50 ( R & D Systems ) . An antigen retrieval step of boiling the sample in 10 mM Sodium Citrate Buffer , pH 6 . 0 for 15 min was used for Oct-3/4 ( C-10 immunofluorescent ) and Chordin staining . Phalloidin staining ( Alexa Fluor , Life Technologies ) was performed according to the manufacturer's instructions . Hematoxylin and Eosin ( Sigma ) staining was performed according to the manufacturer's instructions . Different litters from those used in the microarray analysis were used to confirm changes in gene expression by QPCR . Please see Supplementary Methods ( ‘Text S1 , page 2 ) for assay details . Chimeras were produced as outlined in [72] , and contribution was assessed by semi-quantitative PCR . Please see Supplementary Methods ( ‘Text S1 , ’ page 2 ) for details . RNA was extracted with Trizol according to the manufacturer's instructions ( Invitrogen ) and sent to the UHN Microarray Centre ( Toronto , ON , Canada ) for fluor-labeling ( protocol GE2 v5 . 7 ) , microarray hybridization , and array scanning . Please see Supplementary Methods ( ‘Text S1 , ’ page 4 ) for additional detail and analysis methodology . Please see ‘Text S1 . ’ Please see ‘Text S1 . ’ Please see ‘Text S1 . ’ Please see ‘Text S1 . ’ Please see ‘Text S1 . ’ Please see ‘Text S1 . ’ | Embryogenesis is an intricate process requiring that division , differentiation and position of cells are coordinated . During mammalian development early pluripotent populations are canalized or restricted in potency during embryogenesis . Due to considerable interest in how this fundamental state of pluripotency is maintained , and the requirement of the transcription factor Oct4 to maintain pluripotency , Oct4 has been intensively studied in culture . However , it is not clear what role Oct4 has during lineage specification of pluripotent cells . Oct4 removal during lineage specification indicates that it is required in the primitive streak of mouse embryos to maintain proliferation . The consequences of Oct4 removal diverge from the consequences of removing another factor required for pluripotency between preimplantation development and early cell fate specification suggesting that the network Oct4 acts within is altered between these stages . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [] | 2013 | Oct4 Is Required ∼E7.5 for Proliferation in the Primitive Streak |
Lassa virus ( LASV ) is a significant human pathogen that is endemic to several countries in West Africa . Infection with LASV leads to the development of hemorrhagic fever in a significant number of cases , and it is estimated that thousands die each year from the disease . Little is known about the complex immune mechanisms governing the response to LASV or the genetic determinants of susceptibility and resistance to infection . In the study presented here , we have used a whole-genome , microarray-based approach to determine the temporal host response in the peripheral blood mononuclear cells ( PBMCs ) of non-human primates ( NHP ) following aerosol exposure to LASV . Sequential sampling over the entire disease course showed that there are strong transcriptional changes of the immune response to LASV exposure , including the early induction of interferon-responsive genes and Toll-like receptor signaling pathways . However , this increase in early innate responses was coupled with a lack of pro-inflammatory cytokine response in LASV exposed NHPs . There was a distinct lack of cytokines such as IL1β and IL23α , while immunosuppressive cytokines such as IL27 and IL6 were upregulated . Comparison of IRF/STAT1-stimulated gene expression with the viral load in LASV exposed NHPs suggests that mRNA expression significantly precedes viremia , and thus might be used for early diagnostics of the disease . Our results provide a transcriptomic survey of the circulating immune response to hemorrhagic LASV exposure and provide a foundation for biomarker identification to allow clinical diagnosis of LASV infection through analysis of the host response .
Lassa virus ( LASV ) is a segmented negative-strand RNA virus and a member of the Arenavirus genus . LASV is a human pathogen that is endemic to several countries in West Africa . It is estimated to infect more than 300 , 000 people each year , killing over 3 , 000 with fatality rate for Lassa fever ( LF ) being approximately 15% in hospitalized patients [1] . In several outbreaks 50% case fatality have been reported [2] . LF was initially described as Lassa hepatitis and liver pathology is a significant histological finding in LF patients [3] and in animal models of LF infection [4] . The natural host of LASV is a highly commensal rodent , Mastomys natalensis [5] , [6] , and infection with LASV is thought to occur by direct contact with the host or via aerosol . In part because of its ability to be transmitted through aerosol means [7] , as well as potential for high lethality , LASV has been characterized as a Category A bioweapon agent [8] . There are currently no FDA-approved vaccines or antiviral drugs to treat LASV infection . Ribavirin treatment has been suggested to reduce morbidity in infected patients [9] when initiated within few days of disease onset [10] . Ribavirin is often poorly tolerated and has been associated with a number of severe adverse events . Given the number of side effects associated with this drug , the potential for severe adverse events , and the limited efficacy , there is a strong need for more effective and safer drugs as well as vaccines . In order to develop and test candidate countermeasures , it is critical that there be well-characterized animal models that accurately reflect the human disease . Currently there are several animal models for LASV infection . These include humanized mice expressing human HLA-A2 . 1 instead of murine MHC Class I gene [11] , IFN receptor knock-out [12] mice that can be infected with LASV , and Strain 13 guinea pigs . None of these models completely reproduce the human disease [13] . As a result , more emphasis has been placed on several NHP species that are susceptible to LF [14]–[21] . Both African green monkeys and rhesus macaques show a lethal response to low challenge doses but only a partial response to high challenge doses [14] , [22] . Cynomolgus macaques have been shown to be uniformly susceptible to lethal LASV infection at low and high challenge doses [4] , [16] , [17] , [23] . The development and characterization of these animal models has facilitated the examination of host responses to LASV exposure . In the work presented here , we have used a whole genome microarray-based approach to determine the temporal host response to exposure in PBMCs from cynomolgus macaques following aerosol LASV exposure . Sequential sampling throughout the disease course provided us the opportunity to characterize the circulating immune response to LASV during different stages of infection . Furthermore , we were able to refine this characterization through analysis of immune cell subsets to define the responses of specific effector cells . These analyses showed that there are both rapid and delayed transcription events following LASV exposure , including the upregulation of Toll-like receptor signaling pathways and innate antiviral transcription factors . Our data show that the immune response to LASV involves the expression of a large number of immunosuppressive events in exposed NHPs leading to an inefficient adaptive immune response observed in LASV infections .
Peripheral blood mononuclear cells ( PBMCs ) were isolated from blood pre-diluted with saline using ACCUSPIN System-Histopaque-1077 tubes as per manufacturer's recommendations , and subsequently lysed in TRI Reagent LS ( Sigma-Aldrich ) at USAMRIID . PBMCs were processed for microarray analysis as described earlier [24] . Briefly , total RNA was extracted from the TRI Reagent LS samples , then amplified using the Low-Input Quick Amp Labeling kit ( Agilent ) and hybridized to Whole Human Genome Oligo Microarrays ( Agilent ) in a 2-color comparative format along with a reference pool of messenger RNA ( mRNA ) . Images were scanned using the Agilent High-Resolution Microarray Scanner and raw microarray images were processed using Agilent's Feature Extraction software . The quality of the microarray hybridization for pre-exposure samples obtained from three out of fourteen NHPs in the study was lower than needed for background corrections and normalization . Thus , DNA Microarray data from these three NHPs was not used for data analysis . In summary , out of the 46 samples ( from fourteen NHPs ) hybridized to microarrays , a subset of 30 ( from eleven NHPs ) were used in the subsequent analyses as shown in Figure 1A . The resulting microarray dataset has been submitted to the Gene Expression Omnibus ( GEO ) database , under series record GSE41752 . Subsets of immune cells ( CD4+ , CD8+ , CD14+ , and CD20+ ) were separated from PBMCs by sequential positive selection using nonhuman primate microbeads ( Miltenyi Biotec , Auburn , CA ) as per the manufacturer's recommendations . Although cell separation procedures can cause activation that may affect transcriptional changes , in our study , a number of steps were taken to mitigate these effects . Cells were kept cold , and pre-chilled buffers were used to reduce nonspecific antibody binding , cell surface capping and activation . Additionally , all the pre-exposure ( day −8 ) as well as post-exposure samples were separated using same separation procedure , and all changes in transcription of the pre-exposure samples were subtracted from the post-exposure samples . Therefore , we believe there was little contribution of the separation procedure on the observed transcription . Following incubation with FcR blocking reagent , CD20+ cells were isolated using MS columns and the flow-through fraction was then utilized for the CD14+ isolation . The CD14 flow-through fraction was used for the CD4+ isolation , and the CD4 flow-through fraction for the CD8+ isolation . To increase the purity of the CD20+ and CD14+ magnetically labeled fractions , these were passed over two prepared columns . Aliquots from positive fractions were retained for determination of cell numbers as well as assessment of purity using flow cytometry; the remainder was lysed in TRI Reagent LS . Data were first background-corrected to remove noise from background intensity levels , and afterwards were normalized within the arrays using the Limma package in R [25] . After normalization , the reference and experimental samples were compared to generate log2 fold-change values that represent a change in mRNA expression ( either positive or negative ) . At this step , the internal array control probes were removed . Each array was then further normalized using the pre-exposure control array for that animal to remove monkey-specific expression changes from baseline . A comparison of gene expression was done for day −8 and day 0 samples and there seems to be no difference in the gene expression ( data not shown ) . The resulting dataset was filtered for differential expression and annotated with gene names . The dataset was hierarchically clustered using the Cluster 3 . 0 [26] and visualized using Java Treeview [27] . Functional annotations of gene clusters were assigned using the Database for Annotation , Visualization and Integrated Discovery ( DAVID ) ( http://david . abcc . ncifcrf . gov/ ) [28] . The p-values reported are the value reported by DAVID and are based on the EASE score . The EASE score is an alternative name of Fisher Exact Statistics used in the DAVID system , referring to a one-tail Fisher Exact Probability Value for gene-enrichment analysis . Cytokines were assayed in the plasma of LASV exposed NHPs using a NHP magnetic 23-plex multiplex assay ( Millipore EMD ) in accordance with manufacturer's instructions . Briefly , samples from both pre- and post-exposure time points were assayed in triplicate and washed using a Bio-Rad Bio-Plex Pro II Wash Station equipped with a magnetic manifold . Data were acquired using a Bio-Rad Bio-Plex 3D system and analyzed using Bio-Plex Manager 6 . 0 software and a 5-parameter logarithmic fit . Cytokine levels in assayed samples were derived from the standards run for each assay plate and presented as plasma cytokines in pg/mL . Cytokines/chemokines assayed included granulocyte-macrophage colony-stimulating factor ( GM-CSF ) , interferon gamma ( IFNγ ) , Interleukin ( IL ) -1 beta ( IL1β ) , IL1 receptor antagonist ( IL1RA ) , IL2 , IL4 , IL5 , IL6 , IL8 , IL10 , IL12/23 ( p40 ) , IL13 , IL15 , IL17 , IL18 , monocyte chemoattractant protein-1 ( MCP1 ) , macrophage inflammatory protein ( MIP ) -1 alpha ( MIP1α ) , MIP-1 beta ( MIP1β ) , transforming growth factor-alpha ( TGFα ) , vascular endothelial growth factor ( VEGF ) , sCD40L , and tumor necrosis factor-alpha ( TNFα ) . RNA was isolated from serum of LASV exposed NHPs prepared with TRI Reagent LS ( Sigma-Aldrich ) . The aqueous phase was extracted using Phase Lock heavy gel tubes ( 5 Prime ) and 1-Bromo-3-chloropropane ( BCP , Sigma-Aldrich ) and mixed with 70% ethanol . Following 5 min incubation at room temperature , it was added to an RNeasy column and extracted according to the manufacturer's recommendations ( QIAGEN ) . RNA was eluted through two consecutive additions of 50 µL of nuclease-free water and stored at −80°C until analysis . One-step quantitative real-time RT-PCR reactions were performed on a LightCycler 480 ( Roche , Indianapolis , IN , USA ) in 20 uL volumes with 5 uL of purified RNA and the Superscript II One-Step RT-PCR System ( Life Technologies ) . Primers and probe were specific for the LASV GP gene [Forward , 900 nM: TgCTAgTACAgACAgTgCAATgAg; Reverse , 900 nM: TAgTgACATTCTTCCAggAAgTgC ( Oligos Etc . , Wilsonville , OR ) ; Probe , 200 nM: TgTTCATCACCTCTTC-MGBNFQ ( Applied Biosystems ) ] . Cycling conditions were reverse transcription at 50°C for 15 minutes , and denaturation at 95°C for 5 minutes; then 45 cycles of 95°C for 1 second , 60°C for 20 seconds , followed by a single acquisition; and a final cooling step of 40°C for 30 seconds . Absolute quantification was compared to a viral RNA standard using LC480 software ( version 1 . 5 . 0 . 39 ) and a standard calibrator on each plate . We present viremia data as PFU equivalents/mL using a 10∶1 PCR genome equivalent:PFU ratio that has been previously determined [29] and validated on human LASV samples . RT-PCR assays were carried out to quality check and validate our findings on the DNA microarrays . RT2 Profiler PCR Array from QIAGEN was used to run the RT-PCR . 125 genes of interest were plated on the custom array along with control genes . RNA extracted from the PBMCs ( as described under RNA Processing and DNA Microarrays ) was used for RT-PCR Array . RNA samples collected at three different time points from each animal were run on one plate . In all , six custom plates ( each with a copy of the same probes for 125 genes ) with samples from six different animals were run in this experiment . The RT-PCR experiment was performed as directed by RT2 RNA QC PCR Array Handbook 2012 ( QIAGEN ) . Plates were then run on an ABI 7900 HT qPCR system ( 10 minutes at 95°C , 15 seconds at 95°C followed by 1 minute at 60°C×60 cycles ) . Following the PCR run for all 6 plates , the threshold was made uniform to be consistent among all the plates . Ct values for each sample were obtained . Results were interpreted using SDS software version 2 . 4 and data analysis software from SA Biosciences . Finally , data from different animals on a given day were pooled together and averaged . Data were then presented as fold change over day −8 expressions ( Figure S4 ) .
Prior studies in NHPs have described the course of LASV infection . However , most of these models have focused on disease caused by intramuscular ( IM ) injection . Only one study used aerosol model of LASV exposure [7] . Because human Lassa infection is likely caused by aerosol contact , we were interested in studying the immune response to infection in an aerosol model of exposure . To determine whether the disease progression following exposure to Lassa is similar to studies that carried out exposure to Lassa via IM injection , a pilot confirmation of virulence study was performed in which four NHPs were exposed to LASV via aerosol . At the target dose of 1 , 000 PFU , the actual dose each animal received ranged from 200 to 300 PFU ( Figure S1A ) . Exposed animals all showed signs of disease and succumbed by day 16 post-exposure , with a mean time-to-death of 14 . 5 days ( Figure S1B ) . All four NHPs showed increased levels ( 2–9 fold ) of aspartate transaminase ( AST ) and showed signs of anorexia onset and recumbency at late times ( day 11 onwards ) . These observations and their onset are consistent with low-dose intramuscular LASV challenges in cynomolgus macaques [4] . Animals experienced neurological signs to include seizures ( three of four NHP ) ( Figure S1C ) . Following this confirmation that LASV infection via aerosol exposure led to similar expected clinical signs and disease course development when compared to previous studies , we were interested in determining how circulating immune cells responded to LASV infection . Thus , we participated in a larger study analyzing multiple parameters of infection by sequential sampling of LASV exposed animals throughout the course of disease . A subset of animals was euthanized at different times post-exposure to understand the temporal progression of disease following LASV exposure . From this study , we obtained samples from eleven monkeys ( represented by letters A–K in Figure 1A ) that had been exposed to LASV ( Josiah strain ) via the aerosol route ( target 1000 PFU ) . Prior to challenge , samples were taken for use as pre-exposure baseline controls ( eleven samples at day −8 ) . The samples we obtained were circulating immune cells . At increasing times post-exposure , blood samples were collected and PBMCs were prepared from whole blood . Figure 1A illustrates the distribution of samples in the study . In the table , samples are sorted by days post-exposure . Corresponding clinical observations , viremia , and chemistry data that was also collected is briefly summarized in Figure 1B . The clinical data highlight that there were few signs of disease until 6–8 dpe , when viremia began and increased AST levels began to be observable . Clinical indications as well as severity increased throughout the disease course , with signs of anorexia and initial neurological signs appearing around 10 dpe . Based on this data , we conceptually divided the disease course into three separate stages , early ( pre-symptomatic , 2 to <6 dpe ) , middle ( early symptomatic , 6 to <10 dpe ) , and late ( terminal disease , 10 to 12 dpe ) . These stages and their duration are pictorially illustrated in Figure 1C to highlight their correlation with asymptomatic disease ( early ) , early symptomatic disease ( middle ) , and increasing signs of disease ( late ) . Overall , 30 PBMC samples were processed and hybridized onto DNA microarrays . This resulted in the generation of more than 1 . 3 million data points for analysis . Results from all arrays were computationally analyzed using the Limma software package in Bioconductor , a suite of packages in R . Our initial analysis focused on determining the major changes in mRNA expression over the course of LASV disease . This analysis showed that when experimental arrays were compared to the pre-exposure controls , more than 2 , 000 genes showed at least a 1 . 5 log2-fold change in their expression pattern in at least three arrays ( Figure S2 ) . These genes fall into categories such as immune response ( 153 genes ) , defense response ( 142 genes ) , response to wounding ( 113 genes ) , and inflammatory response ( 83 genes ) , each with a p-value <0 . 00001 . Out of the 2 , 000 differentially expressed genes , 26 . 7% were downregulated and 73 . 3% were upregulated . These genes clustered into several different patterns that were particularly evident when arrays were grouped into categories of early , middle and late disease , similar to the arrangement of Figure 1A . The patterns observed did not appear to be due to the contribution of any one animal , as the removal of multiple arrays followed by re-clustering did not change the patterns observed ( data not shown ) . From this set of 2 , 000 genes that were significantly regulated following LASV exposure , we were particularly interested in the most strongly regulated genes . Figure 2A shows a clustered heatmap of probes that showed a fold-change of greater than 2 . 5 log2 ( >5-fold change ) following exposure . Within these highly regulated probes , three major types of regulation are readily visible in the clustered image . Probes contained in the upper green-boxed region ( expanded in Figure 2D ) showed little change expression early in infection but increased expression late in infection . Probes in the middle box ( expanded in Figure 2C ) showed upregulation at middle and late times post-exposure . Probes in the lower box ( also Figure 2B ) showed strong expression early post-exposure that was maintained throughout the course of the study . Early induced genes were associated with the innate immune response by gene ontology analysis . Genes induced in the middle of the disease course were mRNAs associated with the inflammatory response . Genes strongly induced late post-exposure ( Figure 2D ) were associated with cell cycle regulation , cell division , and DNA packaging . To assess the reproducibility of these findings a number of up and down-regulated genes were assayed by rtPCR ( Figure S4 ) . The selected genes validated well , illustrating that the answers derived from our array analysis are transferrable to a PCR-based detection format . We noted that probes associated with genes involved in the innate immune response markers such as TRAIL ( TNFSF10 ) , MX1 , and CCRL2 demonstrated a rapid innate response that was evident before there are clinical signs of disease ( Figure 2B ) . Early induced genes also included IRF3-responsive genes such as phospholipase A2 gamma , OAS1/2 , and GBP1 [30] . Additionally , we observed that GCH1 showed rapid upregulation . GCH1 is the rate-limiting enzyme regulating the synthesis of pain hormone BH4 and increased GCH1 is associated with increased BH4 levels . Patients with LF often complain of increased pain [31] . Probes that recognized the IL6 gene dominated the list of probes that were upregulated in the middle of the disease course ( Figure 2C ) . IL6 showed little to no induction at early times post exposure but was strongly upregulated by 6 dpe . Several pattern recognition receptors like MARCO , TLR4 , TLR7 , NOD2 , and Fc receptor FCGR2A showed a similar time course of transcriptional activation , as do inhibitory receptors for macrophages function like CMKLR1 , CD200R1 , and CD300LF . Basophil activation marker CD63 was also upregulated beginning around 6 dpe , suggesting the activation of these cells . There was also transcriptional evidence for significant immune cell movement during this stage of LASV disease . At 6 dpe , chemotactic markers such as CCR5 , CCL23 , CXCL12 , and TNFAIP6 were upregulated . This coincident upregulation of innate immune sensors and responses along with repressors of adaptive responses in the form of inhibitor receptors suggests a skewing of the immune response towards the innate . Genes that were upregulated at late times in the LASV disease course ( Figure 2D ) included a large number of granulocyte markers such as TCN1 , CEACAM molecules , and SIGLEC5 . Transcripts of exocytic granule components such as BPI , ELANE and the neutrophil chemotaxis promoting receptor FPR1 , are also upregulated late in response to LASV exposure . Analysis of the cell composition of the blood in the LASV-exposed NHPs at different times post-exposure also suggests the predominance of neutrophils at late times post-exposure ( Figure S3 ) . In addition to upregulation of neutrophil-specific transcripts , we also detected upregulation of immunomodulatory proteins at late times post-exposure: ORM1 , tyrosine kinase genes like BMX and NTRK1 , myeloid specific siglec3 ( CD33 ) , and enzymes such as CHI3L1 , carbonic anhydrase ( CA4 ) , and the apoptosis regulator , NLRC4 . When the strongly upregulated genes from Figure 2A were analyzed using Ingenuity Pathway Analysis ( IPA ) , 65 were observed to be connected through transcriptional or protein-protein interactions ( Figure 3 ) . As was expected from the identification of many innate immune stimulated genes in the early induced population of transcripts , our analysis identified two transcriptional nodes that are upregulated in response to LASV exposure: ( 1 ) IRF3/IRF7 induced genes such as OASs , IFITs , MX1 , and TRIM5; and ( 2 ) STAT1 induced genes such as ISG15 , TRIM25 , GBP2 , TRAIL , and PKR ( EIF2AK2 ) . The identification of both types of genes suggests that the circulating immune cells are generating a strong innate immune response very early post-LASV exposure . Interestingly , there was not a strong transcriptional upregulated interferon α , β or γ in PBMCs , despite the expectation that these genes would be strongly upregulated due to the potential signaling through STAT1 . The identification of IRF3/IRF7 responsive genes in LASV exposure suggested that there was some signaling through either Toll-like receptor ( TLR ) or RIGI-like receptor ( RLR ) . Consistent with this , we saw significant ( more than 1 . 5 log2 fold ) increases levels of several TLR receptors ( TLR1 through TLR7 ) induced at early times post-exposure ( Figure S5 ) . Of the upregulated TLRs , transcripts of TLR3 , TLR4 , TLR5 , TLR6 , and TLR7 were the most strongly expressed . Along with TLR upregulation , we also saw significant increase in the transcripts of RLR genes such as DDX58 ( RIGI ) and DHX58 ( LGP2 ) at early times post-exposure . The coordinated upregulation of these genes strongly indicates that during challenge with LASV the infected host increases its ability to respond through these pathways . IPA analysis also identified a collection of genes that are associated with neutrophil granules . These include three proteases that are upregulated during the course of infection: the matrix metalloprotease MMP9 , the serine proteases neutrophil elastase ( ELANE ) , and cathepsin G ( CTSG ) , as well as the protease inhibitor SERPINB1 . These proteases have important antimicrobial properties , as do other neutrophil granule proteins increased during LASV infections such as lactotransferrin ( LTF ) and lipocalin 2 ( LCN2 ) . This correlates with the increase in the neutrophils seen at late times post-exposure ( percentage and counts , Figure S3 ) in the peripheral blood of LASV-exposed NHPs . This coordinated upregulation of multiple neutrophil granule proteins in the PBMC samples may reflect accelerated granulopoiesis and the mobilization of neutrophil precursors into the blood , consistent with a neutrophillic response to severe LASV infection . Interestingly , protease upregulation has been noted in other animal models of filovirus infection [32] . Moving beyond the analysis of highly differentially expressed genes , we were interested in how the gene expression changes in our arrays compared with changes already noted in animal and human models of disease; therefore , we analyzed the expression of IL1β , IL8 , IL10 , IL6 , IL12 , IFNγ , and TNFα . Earlier reports have followed the expression of these cytokines in LASV-infected animals and humans [4] , [21] , [33] , [34] and have shown that plasma levels of IL8 , IL10 , IL6 , IL10 , TNFα , and IFNγ increase with LASV infection . IL1β was induced very late ( day 13 onwards ) . Our data show that IL1β transcripts are down-modulated following LASV exposure ( Figure 4A and 4C ) . This suppression was clear at early times post-exposure ( animals showing a 1–2 log2 fold decrease ) and persisted throughout the disease course with some variability between animals . Transcripts of IL6 were upregulated during the middle stage of disease , from 6 dpe ( Figure 4A and 4C ) . Transcripts of IL8 were not differentially expressed in PBMCs at any stage ( early , middle or late ) of LASV disease . Similarly , IL10 transcripts were not differentially expressed in the PBMCs of LASV-exposed NHPs , nor were IL12 transcripts . These results are in line with earlier reported studies on fatal cases of LASV patients [33] . Transcripts of IFNγ were down , and transcripts TNFα were not differentially expressed in our model of LASV infection; however , during the middle stage of LASV disease , there was an upregulation of the TNFα-induced genes , TNFAIP2 and TNFAIP6 ( Figure 4B and 4C ) . This finding correlates well with an earlier study on human patients of LF [35] . In addition to this data showing concordance of our data with previous studies , we also saw differential expression of cytokines that have not been previously reported in the context of LASV infection . Transcripts of IL21 were slightly upregulated ( 1–1 . 5 log2 fold ) in PBMCs at 10 dpe ( Figure 4D and 4E ) . Transcripts of IL21 were strongly upregulated ( more than 3 log2 fold ) in the separated CD4 positive cells ( Figure 4G and 4H ) . Upregulation of IL21 has been reported in LCMV model of HF and has been linked to the clearance of infection but had not previously been noted in LASV infection [36] . Transcripts of IL23α and IL24 were down-modulated ( 2–1 . 5 log2 fold ) at 3 dpe and 6 dpe respectively , during LASV exposure ( Figure 4D ) . Interestingly , we detected upregulation ( 1 . 5 log2 fold ) at 6 dpe of transcripts of an immunosuppressive cytokine , IL27 ( Figure 4D and 4E ) . This upregulation of IL27 was more pronounced ( 4 log2 fold ) at 6 dpe in CD4 positive cells than in the PBMC population ( Figure 4G and 4H ) . IT is possible that IL27 protein may play an important function in the response to LASV infection . In addition to these changes in cytokine expression , chemokines such as CCRL2 and CCL23 are also upregulated in the LASV-exposed NHPs . CCRL2 , a marker of monocytic in-filtration in inflammatory diseases [37] , is significantly upregulated ( 1 . 5–2 log2 fold ) early ( 3 dpe ) following LASV exposure , and CCL23 [also known as macrophage inflammatory protein 3 ( MIP3 ) ] , a potent chemoattractant for T lymphocytes and monocytesis , is upregulated ( more than 3 log2 fold ) at middle times post-exposure ( 8 dpe ) ( Figure 4F ) . Together , the upregulation of CCRL2 and CCL23 suggests that LASV exposure causes an increase in the call for the recruitment of inflammatory cells . To understand the relation between the timing of transcript upregulation in PBMCs and the onset of viremia in LASV disease , viral load was evaluated throughout the course of infection by RT-PCR . Transcripts a selected set of upregulated genes were compared to the viral load in the plasma at different time points in Figure 5 . Comparisons in Figure 5A highlights that while viremia was not observed until 8 dpe , transcripts of cytokine IL27 showed increased expression earlier ( approximately 6 dpe ) . More dramatically , innate antiviral response molecules such as IRF7 , STAT1 , and IFIT2 , and chemokines such as CXCL12 are upregulated by 3 dpe ( Figure 5B ) . This demonstrates that the expression of host transcripts ( immune response ) in response to LASV exposure significantly precedes the onset of circulating viremia in infected NHPs . We also compared the cytokine gene expression to the cytokine protein expression seen in the plasma of LASV exposed NHPs ( Figure 6 ) . Cytokine levels in blood samples were determined prior to LASV exposure and throughout the course of disease . This comparison of gene expression with the actual protein levels of these cytokines ( Figure 6 ) revealed that the increased gene expression of a subset of cytokines correlated with protein expression ( e . g . IL8 , IL6 and IL18 ) . However , this correlation was not always observed . For IL8 , an initial correlation of mRNA and protein expression at early times broke down by 12 dpe , where we observed a loss of mRNA expression but protein expression was still readily detectable . This could be due to either loss of the IL8 producing cell population from the PBMCs due to migration or cell death , or it is also possible that another cell population accounts for the observed protein in the plasma at late times of disease . We also observed that array analysis correctly characterized proteins that were expressed in response to viral exposure , such as IL1β , IL12 , and IL10 ( data not shown ) . As LASV disease in humans is associated with an inefficient host immune response to infection [38] , we determined the expression level of several immune suppressive genes ( negative regulators of immune responses ) with the hypothesis that PBMCs would upregulate immunosuppressive genes would be upregulated in PBMCs in response to LASV exposure . Table 1 describes the upregulation of genes involved in regulating T cell function , such as PDL1 ( CD274 ) [39] and TNFRSF21 [40] , that are upregulated within the first 6 dpe . We also see upregulation of the regulatory players that are involved in intracellular signaling and have been linked to immunosuppressive function . For example , transcripts of the following were upregulated during LASV disease: ITIM domain containing receptor LILRB2 , an adapter protein DOK3 [41] , [42] , an acute phase protein ORM1 [43] , an ADP ribosylation factor PARP14 [44] , a phosphogluconate dehydrogenase PGD [45] , a translational repressor SAMD4A , a neutrophil elastase inhibitor SERPINB1 [46] , and TGFβ induced protein TGFBI . TGFBI can cause decrease the p53 binding to DNA leading to down-modulation of p53 mediated genes ( DR5 , JUNB , TET2 , and GADD45 , as seen in our gene expression data ) . TGFβ has been reported in the plasma of LCMV-infected rhesus macaques [34] . Along with upregulation of these immunosuppressive genes , there was a significant early down-modulation of positive immune response regulating genes such as JUN , FOS , CD69 , CD83 , STAT4 , GADD45A , and SIGLEC10 in PBMC from LASV exposed NHPs ( as shown in Table 1 ) , suggesting the lack of key players in the generation of immune response to infection . Expression of these genes is shown as a heatmap and line graphs in Figure S6 . Together , the behavior of these genes is consistent with an upregulation of immunosuppressive signals , compounded by the lack of factors that lead to the development of adaptive immune response such as production of IL2 and IL4 following virus infection .
At the outset of this study , it was unclear whether the PBMC population would provide an accurate report of the systemic immune response; however , our analysis has proven that these cells appear to respond to LASV exposure in ways that closely mirror what has already been described for the response to infection in humans and NHP [4] , [21] , [33] , [47] . These changes include the lack of IL1β , IL8 , TNFα , IL2 , IFNγ , IL4 , IL12 , and IL10 , and the upregulation of IL6 . In our system , we see an induction of IL6 , consistent with earlier descriptions of LASV associating this cytokine with fatal cases of human patients [48] as well as animal models [4] . Our data show that there is both an early induction as well as sustained activation of IRF3/IRF7 responsive genes following aerosol exposure to LASV ( Figure 2 ) . This robust response was evident very early post-exposure prior to onset of clinical signs or detection of viremia and was maintained throughout the disease course . This potent innate response does not appear to be coupled to a strong adaptive immune response , as IL4 , IFNγ , and IL2 were not upregulated in the PBMC at any point in response to LASV exposure . One suggestion for why there is a poor adaptive response is that immunosuppressive responses are dominating the PBMC response to exposure . Our data shows that there is strong expression of genes which can have interferon-suppressive activity , such as TGFBI [49] , CD274 ( PDL1 ) , TNFRSF21 , and IL6 [50] . This is consistent with the hypothesis that the overwhelming innate antiviral response seen in LASV-exposed macaques is compromising the downstream immune response by either inhibiting DC differentiation [51] or cytokine responses [52] , or causing attrition of T cells ( mainly CD8+ cells ) [53] , [54] . We also found that during LASV disease there is increased expression of IL27 and IL21 mRNAs . Upregulation of IL21 was previously noted in a LCMV model of hemorrhagic fever [36] and recent studies have provided genetic evidence that the IL21 gene is important for successful responses to LASV exposure [55] , suggesting that further analysis of the importance of IL21 in LASV exposure is warranted . Our finding that IL27 transcripts are upregulated early in LASV exposure is interesting because IL27 enhances the expression of the immunosuppressive PDL1 gene [56] and inhibits T cell function . Thus , increased IL27 provides a potential mechanism for the observed suppression of adaptive cytokines such as IL2 and IL4 . Comparison of our results with other microarray-based analyses of arenavirus infection shows that our data correlate well with other models of arenavirus-induced hemorrhagic fever [34] . Both studies found similar changes in the STAT1/interferon responsive genes such as TRAIL , IFI44 , OASs , IFIT1 , and IFIT2 showing a strong innate immune response . Our data suggest that the innate response is slightly more rapid in LASV than in LCMV infected primates . Both studies show similar changes in the down-modulation of immune response genes such as IL1β , CXCR4 , IL8 , and IL24 and the activation of transcription factors STAT1 , STAT2 , SOX4 , NR4A2 , SMAD7 , and TRIM25 . Thus , comparison of our study with LCMV infection model suggests that arenavirus infection can lead to highly similar transcriptional fingerprints , which have a few notable differences in differing kinetics of interferon induction and expression of CCL5 and CREB1 . The similarity of the transcriptional response in the two arenavirus models of hemorrhagic fever is in stark contrast to that seen in Ebola virus ( EBOV ) -infected animals . In both EBOV- and LASV-exposed animals , there is a strong induction of IRF3 and STAT1-responsive genes [24] , [57] , but the response following LASV exposure appears to be earlier than that seen for EBOV-infected monkeys . Also , a lack of IL1β and low levels of IL8 are seen in the LASV model , but these genes are strongly expressed in cells from EBOV-exposed NHPs [24] , [57] . PBMCs from LASV exposed animals do not express the apoptotic and clotting factor genes associated with EBOV disease , further highlighting the fundamental differences in the immune response to these different infections . Our studies suggest that tracking gene expression in circulating immune cells is a worthwhile avenue to explore for early disease diagnosis . Comparison of cytokine gene expression with the amount of detectable protein in LASV-exposed NHPs demonstrated a positive correlation between the gene expression and protein expression for a subset of cytokines such as IL6 , IL8 , and IL18 , suggesting no diagnostic or prognostic advantage for transcript-based analysis . In contrast , our microarray system provided the ability to identify major changes in interferon-stimulated genes considerably before the onset of viremia in LASV exposed NHPs ( Figure 5 ) , suggesting that analyzing the expression of these transcripts or proteins could provide pre-symptomatic detection of disease . This would be directly analogous to recent work showing that pre-symptomatic detection of influenza is possible [58] but additional analysis will be necessary to fully assess the possibility . To our knowledge , this is the first whole-transcriptomics analysis of the response of the primate circulating immune system to LASV exposure . This analysis validates the underlying model as a faithful reproduction of the human disease and demonstrates that much of the immune response can be tracked through analysis of the circulating immune cells . Our results highlight that a transcriptomics approach allows the analysis of previously investigated cytokines such as IL1β , IL8 , and IL6 . Additionally , it allows the simultaneous analysis of additional genes that this study has found are strongly upregulated , such as IL27 and CD274 ( PDL1 ) . Our data also shows that the circulating immune response shows a strong innate response to infection that is visible hours to days before the earliest clinical assay ( viremia ) identifies infection in this model , suggesting that future studies may be able to capitalize on this information to develop pre-symptomatic diagnostics for LASV infection . | Lassa virus ( LASV ) , a member of the Arenaviridae family , is a viral hemorrhagic fever causing virus endemic to several countries in West Africa with a history of sporadic importation into the United States . It has been characterized as a Category A agent , and despite the significant public health issues posed by LASV and the potential biodefense risks , little is known about the immune response to the virus . In the study presented here , we have taken an unbiased genomics approach to map the temporal host response in the peripheral blood mononuclear cells ( PBMCs ) of non-human primates ( NHP ) exposed to LASV . Gene expression patterns over the entire disease course showed that there are strong transcriptional changes of the immune response to LASV exposure , including the upregulation of Toll-like receptor signaling pathways and innate antiviral transcription factors . However , there was a lack of pro-inflammatory cytokine response in LASV exposed NHPs similar to what is seen in human disease . Our data suggests that LASV induces negative regulation of immunological events , leading to an inefficient adaptive immune response as observed in LASV-infected human patients . Our results provide a picture of the host's circulating immune response to hemorrhagic LASV exposure and demonstrate that gene expression patterns correlate with specific stages of disease progression . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] | [
"immune",
"activation",
"immunology",
"host-pathogen",
"interaction",
"microbiology",
"animal",
"models",
"adaptive",
"immunity",
"model",
"organisms",
"immune",
"defense",
"immunoregulation",
"animal",
"models",
"of",
"infection",
"biology",
"immune",
"response",
"macaque",
"immunity",
"virology",
"innate",
"immunity"
] | 2013 | Transcriptional Profiling of the Circulating Immune Response to Lassa Virus in an Aerosol Model of Exposure |
The genes involved in conferring susceptibility to anxiety remain obscure . We developed a new method to identify genes at quantitative trait loci ( QTLs ) in a population of heterogeneous stock mice descended from known progenitor strains . QTLs were partitioned into intervals that can be summarized by a single phylogenetic tree among progenitors and intervals tested for consistency with alleles influencing anxiety at each QTL . By searching for common Gene Ontology functions in candidate genes positioned within those intervals , we identified actin depolymerizing factors ( ADFs ) , including cofilin-1 ( Cfl1 ) , as genes involved in regulating anxiety in mice . There was no enrichment for function in the totality of genes under each QTL , indicating the importance of phylogenetic filtering . We confirmed experimentally that forebrain-specific inactivation of Cfl1 decreased anxiety in knockout mice . Our results indicate that similarity of function of mammalian genes can be used to recognize key genetic regulators of anxiety and potentially of other emotional behaviours .
Exploiting naturally occurring genetic variation to identify mechanisms that give rise to behavioural phenotypes in mammals has proved to be extremely difficult [1] . The abundance and small size of loci that contribute to behavioural variation frustrate gene identification and make it difficult to know which among them are central to the responsible biological mechanisms [2] . A major challenge is to devise methods that move quickly from locus to mechanism [3] . Using heterogeneous stock ( HS ) mice descended through more than 50 generations from eight inbred progenitor strains [4] we have previously identified 205 quantitative trait loci ( QTLs ) that contribute to variation in one or more of four anxiety tests: the elevated plus maze , open-field arena , freezing to the context and reluctance to try a novel food . Performance levels on these tests reflect , at least in part , activity in the ventral hippocampus [5] , [6] , [7] , [8] , [9] and these tests were chosen in order to interrogate an underlying psychological construct of anxiety from different perspectives ( a single measure , such as variation in locomotor activity in the open-field arena , will include traits irrelevant to anxiety [10] ) . We set out to determine causal genes for anxiety in the HS . Recombinants that have accumulated since the founding of the HS means that QTLs are mapped to intervals of approximately 3 Mb [4] , much higher resolution than obtained by mapping in backcrosses or intercrosses [3] . Nevertheless , mapping in the HS rarely identifies single genes so additional approaches are necessary to identify candidate genes . Here we considered an alternative approach based on two assumptions . The first is that the mosaic structure of the genomes of inbred laboratory mice could be used to reduce the regions containing candidate genes . Because HS mice are descended from a small number of founders , any pair of mice will share a fraction of their genome [11] so that all genomic regions can be classified as either identical or non-identical by descent . Since a QTL must lie in a region where sequence differences distinguish strains in the same way as the QTL alleles , in a cross between two strains the QTL must lie in a region that is not identical by descent . In multiple crosses , or in animals descended from multiple strains such as the HS , this relationship , though more complex , still holds and can be used to fine-map QTLs [12] , [13] . The second assumption is that genes that influence the same , or related , traits have similar functions which are captured by existing functional annotations [14] . The gene ontology ( GO ) database , for example , assigns biological descriptors ( GO terms ) to genes [15] . Genes assigned the same GO term can be regarded as members of a category of genes that are more closely related in terms of some aspect ( s ) of their biology than are randomly-chosen genes . Therefore the presence of a highly non-random pattern of functional annotations is an indication that we have correctly identified genes influencing a trait . Importantly , we do not make any assumptions about which annotations are relevant to a trait prior to performing our analysis .
We used 205 QTLs that contribute to variation in four different anxiety tests: the elevated plus maze , open-field arena , freezing to the context and reluctance to try a novel food . The identification of these QTLs in HS mice is described in [4] . Our first aim was to determine regions within QTLs that are most likely to contain genes involved in the phenotype . To do so , we began by dividing up the genomes of the HS progenitors according to the pattern of ancestral allelic similarities and differences at a locus . Our intention was to identify regions of the genome descended from a common progenitor . Just as each HS mouse is descended from eight inbred strains , the progenitors in turn have ancestors in common . Using a dynamic programming algorithm we partitioned the genome into regions in which all sequence variants detected in the near-complete genome sequence [16] are consistent with a single phylogenetic tree [17] . We next used the phylogenetically determined strain distribution patterns to find regions likely to contain genes with variants that could be causally related to phenotypes using a merge analysis [13] . On most phylogenetic trees some founder strains are indistinguishable and so share the same leaf: in other words , the tree merges the eight strains into groups that share alleles . Causal variants lie in intervals where the tree partitions strains consistent with the allelic effects of the QTL . We refer to these intervals as consistent QTL intervals . The 205 anxiety QTLs include 5 , 932 genes ( 29 genes per QTL ) , while the consistent QTLs contain 458 genes ( 2 . 4 genes per QTL ) . Figure 1 presents an example of a merge analysis for a QTL on chromosome 19 . Reasoning that causal genes within QTLs for a specific trait are likely to share functions we looked for enrichment of functional annotations ( gene ontology ( GO ) annotations ) in the 458 genes within the consistent QTLs . Many tests of functional enrichment assume that the functions of neighboring genes in consistent QTLs are uncorrelated . However neighbouring genes may have similar functions , as tandem duplications , which occur throughout the genome , often give rise to functionally related genes [18] . We addressed this problem using a permutation test that accounts for gene order within genomic intervals . The test assigns P-values to GO annotations , representing the strength of the evidence against the null hypothesis that GO annotations are randomly distributed amongst consistent QTLs . We summarised the functional coherence of GO annotations associated with the set of genes within consistent QTLs . We define the ‘information score’ as the sum of the negative base-10 logarithms of the P-values ( logP ) for all GO terms associated with genes within the intervals . The information score can be considered to be a measure of the degree of coherence within the set of genes compared to that in a random sample of genes , and has similarities to the self-information measure of information theory . For the enrichment analysis we combined genes from all phenotypes , since our aim is to identify biological features that reflect the underlying psychological construct of anxiety , rather than to identify test specific features . The location of QTLs for different measures of anxiety sometimes coincides , for example when we have multiple measures from a single test , such as the open-field arena and elevated plus maze . For overlapping QTLs , we included the QTL with the smallest 95% confidence interval . Figure 2A shows a significant enrichment in information score for the genes in the consistent QTL intervals compared to values calculated from 10 , 000 sets of randomly sampled genes for each of the trait sets . Figure 2B shows that no significant enrichment was found when an identical analysis was performed using all genes at each QTL , ignoring the results of the merge analysis . We tested for over-representation of GO terms , since it is not clear how to validate genes associated with under-represented GO terms ( an under-represented gene would be one that is not involved in the phenotype ) . At a 10% false discovery rate ( FDR ) we identified 16 GO terms that are over represented in 167 genes at 57 QTLs ( Table S1 ) . More than 90% of the genes were identified by domain-level terms ( biological and cellular process ) or high-level terms ( anatomical structure development , system development , developmental process , multicellular organismal process , cellular metabolic process ) . Only three GO terms yielded information about specific mechanisms: two genes ( cofilin-1 ( Cfl1 ) and destrin ( Dstn ) ) were associated with “positive regulation of actin filament depolymerization’ , and a single gene was associated with both “eye pigment granule organization and biogenesis” and “lens morphogenesis in camera-type eye” . However this gene , Shroom2 , is also associated with the GO term “negative regulation of actin filament depolymerization” , suggesting that actin filament depolymerization might be an important mechanism involved in anxiety . We examined the effect of disrupting polymerisation/depolymerisation of actin filaments in the hippocampus of mice by using a conditional mutant , n-Cofflx/flx , CaMKII-cre , in which Cfl1 is deleted in the principal neurons of the developed forebrain ( which includes the hippocampal formation ) [19] , [20] , [21] . Previous work has demonstrated that CamKIIα-cre mice are indistinguishable to wild-type littermates ( e . g . [22]–[23] ) so we employed littermate mice as controls for the experiments described below . Anxiety occurs when there is a conflict between competing goals or response options [24] , [25] . For example , most unconditioned laboratory tests of anxiety rely on the conflict between whether the animal should approach and explore the relatively more open and exposed sections of the apparatus , or avoid these potentially more dangerous areas . Changes in approach/avoidance behaviour in novel , mildly aversive environments were used as a measure of anxiety , and are dependent , in part , on the ventral hippocampus [5] , [6] , [7] , [8] . An anxious rodent will be slower to enter , and will spend less time in , the more open and exposed sections of the apparatus ( e . g . open field arena ( OFA ) and elevated plus maze ( EPM ) ) . They will also defecate when placed in a brightly lit OFA [6] . Numerous studies have used anxiolytic drugs to show a correspondence between the behaviour of rodents in the OFA and EPM and human anxiety [26] , [27] . In both the OFA ( Figure 3 ) and EPM assays ( Figure 4 ) we found that the Cfl1 mutants were significantly less anxious than controls . Mutants showed significantly increased total activity ( Figure 3A ) and decreased latency to enter the central region in the OFA ( Figure 3C ) . They also defecated less during the OFA test ( Figure 3B ) . The Cfl1 knockouts also spent significantly more time in the open arms of the EPM ( Figure 4D ) . They had longer path lengths within the open arms ( Figure 4C ) , had a reduced latency to enter an open arm for the first time ( Figure 4E ) , and made more entries/visits into the open arms ( Figure 4A ) . Importantly , however , the number of entries into the closed arms of the EPM did not differ between the groups , suggesting that these changes in behaviour do not simply reflect a generalized locomotor hyperactivity in the Cfl1 knockouts ( Figure 4B ) . To explore this further , we examined locomotion of single-housed mutant mice under stress-free conditions using infrared sensors to detect spatial displacement over time in standard mouse cages . In a 24 hour period , neither total activity , nor activity during the light or dark phases , were significantly different between the two genotypes ( Figure 5 ) .
In this paper we show how the near-complete sequence from the progenitors of the HS can be use in conjunction with gene annotations to identify genes influencing anxiety at QTLs in HS mice . The method we applied involves partitioning QTLs into intervals that can be summarized by a single phylogenetic tree among the HS founders , testing whether that partitioning was consistent with alleles influencing anxiety at each QTL , and then searching for common functions in candidate genes positioned within those intervals . Crucially , we were able to show there was no enrichment for function when we included all genes under each QTL , thus confirming the value of phylogenetic filtering . Our method is a development of two analytical techniques , probabilistic ancestral haplotype reconstruction ( HAPPY ) [28] and merge analysis [13] , but it is not a replacement for either; rather , it depends on both . HAPPY is a tool for mapping in populations whose progenitors are known ( or can be inferred ) , while merge analysis identifies which variants might be functional , based on a comparison between the HAPPY derived allelic effects and those of the variant . Incorporating phylogenetic filtering into merge analysis allows us to determine which regions ( rather than which variants ) are putatively functional and hence to prioritize genes that lie in these intervals for functional studies . Phylogenetic filtering is the methodological advance described here . Our approach has some obvious limitations . Above all , the relevant gene at most QTLs still remains unknown . At best , we identified genes at 57 QTLs out of 205 . Even allowing that the same QTL influences multiple measures , genes at more than half of the QTLs are not identified . This may in part reflect our reliance on an imperfect set of annotations . As the quality and density of annotations increases , it may be possible to detect more functional patterns among genes at consistent QTLs . However failure to find enrichment may also reflect a problem inherent in all sequence based approaches: finding functionally relevant sequence does not immediately translate into finding functionally relevant genes . Typically , as here , genes are identified because they either contain , or lie close to the functionally relevant sequence , but proximity does not unequivocally indentify the correct genes . We were also unable to find many terms that pointed to a potential mechanism . Again this likely reflects the relative poverty of annotations . Despite these limitations , we think our method has important advantages . Notably it addresses an emerging problem in mouse complex trait , namely the need to prioritize large numbers of candidate genes . Until recently there were relatively few loci mapped at sufficiently high resolution to suggest high quality candidate genes for functional studies . The use of resources that can deliver near gene-level mapping resolution ( HS mice , commercial outbreds [29] or the Collaborative Cross [30] , [31] ) , together with the realization that hundreds , if not thousands , of individual genetic variants are involved [32] , is about to transform that situation . A critical problem for mouse complex trait analysis problem now is how to validate the large number of candidate genes the new mapping resources identify . The many genes we identified by searching for enrichment of domain-level or high-level GO terms likely provide a useful starting point for functional studies . It should be noted that they include a number of ion channels and neurotransmitter receptors ( see Table S1 ) . Two further observations are worth making about the use of sequence for the identification of Cfl1 as a quantitative trait gene . First , the sequence variants contributing to the QTL likely lie in a regulatory region . From the available progenitor strain sequence we know that no sequence variants segregate in the HS within the Cfl1 gene itself . The nearest 5′ variant is a SNP at 5 , 489 , 197 ( the transcriptional start site of Cfl1 is at 5 , 490 , 455 ) and the nearest 3′ is a SNP at 5 , 494 , 237 ( the end of the gene is annotated as 5 , 494 , 031 ) . Second , previous mapping of transcript abundance in the HS identified a ci-acting expression QTL for Cfl1 ( in the hippocampus ) with a logP of 26 . 4 and a peak at approximately 5 . 8 Mb ( [33] see http://gscan . well . ox . ac . uk/gsBleadingEdge/wwwqtl . cgi ) . It is possible that the variants contributing to the expression QTL are also those that contribute to the behavioural phenotype ( unfortunately we cannot determine whether the alleles in the HS act in the same direction as in the knockout experiment , due to the correlated nature of allelic effects in the HS [4] ) . A second issue that warrants discussion relates to the importance of what we have found , namely a relationship between actin filament depolymerisation and genetic differences in anxiety behaviour in the mouse . Since the method depends on gene annotations , we face the objection that we are limited to the discovery of what is already known . Does our work represent an advance in understanding the biology of anxiety ? Rust and co-workers have previously shown that Cfl1 plays an important role in controlling dendritic spine morphology and that the n-Cofflx/flx , CaMKII-cre mice are deficient in long-lasting forms of synaptic plasticity when assessed using hippocampal slices [21] . n-Cofflx/flx , CaMKII-cre mice display behavioural impairments in long-term associative spatial memory [21] , as shown by impairment on the standard , spatial reference memory version of the Morris water maze task , in which mice were required to form a long-term association between a particular spatial location and the presence of the escape platform . This raises the possibility that differences in spatial memory abilities and in spatial exploration could have contributed to the observed differences between n-Cofflx/flx and n-Cofflx/flx , CaMKII-cre mice in both the OFA and EPM in the present study . Against this it is important to point out two things . First , despite their inability to form long-term associative spatial memories , the n-Cofflx/flx , CaMKII-cre mice displayed normal performance on tests of short-term spatial memory [21] . This suggests that the Cfl1 knockout mice are able to discriminate between spatial locations perfectly well , and to acquire this information rapidly . It also shows that these mice do not have a general problem with all aspects of spatial information processing . Second , lesions of the ventral hippocampus have no effect on spatial learning and memory performance . In contrast , lesions of the dorsal hippocampus impair spatial learning and memory but have no effect on tests of anxiety [34] . This double dissociation between the effects of dorsal and ventral hippocampal lesions suggests that the hippocampus may have multiple , dissociable functions associated with different sub-regions of the hippocampus , and that changes in anxiety levels in the Cfl1 knockout mice are unlikely to be due to differences in spatial memory abilities or spatial exploration . We have assumed here that the effects we observe in the transgenic animals are due to genetic ablation restricted to the hippocampus , but we cannot exclude the involvement of the amygdala . While expression of the Cre recombinase occurs predominantly in the pyramidal neurons of the hippocampus , it also occurs in the striatum , and amygdala [23] . The latter structure is also involved in mediating emotional behaviours , although there appears to be some division of labour between hippocampus and amygdala [35] The anxiety tasks associated with ventral hippocampal lesions are the approach/avoidance tests used here [8] , [34] , [36] which it is worth noting are generally unaffected by lesions of the amygdala . Thus we argue that dysfunctional cytoskeletal remodeling and the consequent alterations in synaptic plasticity in the hippocampus , and particularly the ventral hippocampus , are the most likely mechanism that contributes to the altered anxiety levels in Cfl1 knockout mice . Cytoskeletal remodelling is linked to synaptic plasticity and synaptic plasticity is a key neural substrate for emotional behaviours , including anxiety . Indeed , NMDA receptor-mediated synaptic plasticity in the hippocampus is a key determinant of anxiety levels [37] . Anxiety-like states in rodents [38] , [39] and humans [40] alter hippocampal dendrites , presumably reflecting synaptic changes . In vertebrates , excitatory synapses are found predominantly on dendritic spines where actin is highly enriched and provides the structural foundation for changes associated with postsynaptic specialization [41] . Electrophysiological measures of synaptic plasticity , long-term potentiation and depression have been associated with growth and shrinkage of dendritic spines respectively [42] , [43] , [44] . Disruption of genes involved in spine formation can also cause deficits in anxiety behaviour [45] and it has been shown that Lipocalin-2 ( Lcn2 ) regulates stress-induced anxiety in mice via changes in spine morphology and density [46] . Our findings add to this growing literature on the relationship between anxiety-like behaviour and alterations in dendrites . Cfl1 is likely to affect anxiety via the hippocampus , and more specifically the ventral hippocampus . To our knowledge this is the first time that Cfl1 has been implicated as a gene influencing anxiety-like behavior .
To take into account the differing degrees of relatedness in the HS we use a mixed models approach where a covariance matrix of the genetic random effects quantifies relatedness in the HS . Variance components were estimated using the R package EMMA [47] . To compare the fit of the strain distribution pattern to the genetic action of the QTL we applied a statistical test , called merge analysis [13] . Merge analysis is related to imputation methods used in human GWAS . It tests whether the strain distribution pattern sequence variants across the HS founders is consistent with the estimated trait values for the founders , by comparing the fit of a QTL linear model in which each founder strain can take a different trait value to one in which those founders sharing the same DNA variant allele are merged and constrained to take the same value . The merge statistic is the negative logarithm of the P-value ( logP ) of the ANOVA of the merged model . When this value equals , or exceeds the logP of the unmerged model ( the unconstrained 8-way haplotype test [28] ) the DNA variant could be a QTL allele . We applied merge analysis with one modification . Our aim was to determine regions likely to contain genes involved in the phenotype , rather than identify the causal variant . So , within the 95% confidence interval for each QTL , we segmented the locus into intervals between the SNPs in the HS mice , based on the ancestral recombination graph among the eight HS progenitors . Within each interval all sequence variants detected in the Sanger mouse genomes database [16] are consistent with the same ancestral tree ( i . e . every pair of variants obeys the 4-gamete test ) . This method is described in [17] . On most trees some founder strains are indistinguishable so share the same leaf . Therefore we used the tree to represent each interval in the merge analysis , by generating a pseudo multi-allelic marker whose alleles correspond to the leaves , and comparing the fit of the tree to the 8-way haplotype test . We designate a merge interval as one whose logP value equals or exceeds the logP of the haplotype test . Thus the merge intervals act as an importance filter on the QTL intervals , subdividing each QTL into regions that could contain causal variants , and therefore are more likely to contain the causal genes . The test was coded in R as an extension to the R HAPPY package ( http://www . well . ox . ac . uk/rmott/happy ) . If the merge intervals were more likely to contain causal genes than the QTL intervals as a whole , we would expect them to be enriched for certain classes of genes . We tested for over-representation of gene function annotations within the merge intervals . Our null hypothesis is that the merge analysis places intervals randomly within QTL intervals , rather than correctly identifying causal variants for the trait being investigated . However we have to take into account a number of potential biases . First , there may be a bias due to chromosomal location or G+C content of the QTLs . Our sampling procedure therefore draws sampled intervals matched both for chromosome and G+C content . Second , enrichment in GO descriptors could simply be due to larger numbers of genes found within the intervals . We used a procedure that matched the gene numbers within the random intervals to that of the QTL intervals to avoid this type of bias . Finally , tests that assume independence , such as the hypergeometric test , may not provide robust estimates because neighbouring genes within a QTL may have similar functions ( for example , functionally similar genes arising from tandem duplications ) . We employed a Monte Carlo simulation method to test for over-representation of gene function annotations within genomic intervals . This method does not assume that the function of each gene sampled is independent . The Monte Carlo simulation first identifies all genes that overlap QTLs . For each biological process GO annotation term j , we counted the number of QTLs ( nQj ) that overlap any of the genes associated with that annotation . To identify GO terms that are significantly enriched among genes within consistent QTLs , we created a null distribution from 5 , 000 sets of randomly sampled genomic intervals , each with a length distribution identical to that of the test set of QTLs . Each set was drawn from regions of similar nucleotide ( G+C ) content . Chromosomes were divided into 1 Mb-sized windows and each window assigned to one of 10 equally populated %G+C bins . For each of the test QTLs we picked a random genomic location from the same chromosome that was located in a genomic window from the corresponding %G+C bin . Each randomly sampled interval was overlaid with a set of simulated consistent QTL intervals identical to that of the test QTL . For each GO term j , the number of randomly sampled regions nrj that overlap genes associated with j was calculated , ignoring genes outside the simulated consistent QTLs . The fraction of these 5 , 000 sets for which nrj≥nQj is pj , which represents an estimate of the probability that annotation j is observed in nQj QTLs simply by chance . A further 5 , 000 sets of randomly sampled regions ( defined as above ) were used to determine the experimental false discovery rate ( FDR ) . For each set , the number of significantly over-represented annotation terms was recorded . The P-value threshold giving the desired FDR value was then applied to the results for downstream analysis . Gene targeting of the Cfl1 gene was performed in 129 Sv embryonic stem cells [19] , [20] . The conditional Cfl1 allelle was backcrossed onto a C5BL6/J background for more than 20 generations . Inactivation of Cfl1 in the principle neurons of the adult forebrain was achieved by crossing a CamKIIα-cre transgene onto the conditional Cfl1 strain [20] , [21] , [23] . Behavioral analyses were performed on male mice using age-matched littermates ( n-cofflx/flx ) as controls . A first cohort of 6–7 week old mice was tested first in the open-field and next in the elevated plus maze . A second cohort of 8–10 week old mice was used for activity recording in a home cage . Mice were housed in an animal facility with 12-hour light-dark cycle and water and food access ad libitum . Animal treatment and care were provided in accordance with institutional guidelines . Home cage locomotion of single-housed mice was assessed in standard mouse cages ( Type II ) using TSE InfraMot infrared sensors ( TSE Systems , Bad Homburg , Germany ) . Mice were transferred to new cages 12 hours before starting the recordings . Anxiety was assessed in mice using two different , ethological , unconditioned tests of anxiety . These were the open field arena ( OFA ) and the elevated plus maze ( EPM ) . A standard rectangular OFA with 0 . 5 m side length ( TSE Systems , Bad Homburg , Germany ) was used . At the beginning of the experiment , mice were placed in one of the corners facing away from the center region . Locomotor activity and latency of entering the center region were assessed using the VideoMot2 video tracking system ( TSE Systems , Bad Homburg , Germany ) . Fecal boli were counted manually at the end of each experiment . A T test was used to compare locomotor activity; the Mann-Whitney test was used to compare fecal boli counts and the Mantel-Haenszel test implemented in the ‘survival’ package was used to compare latency of entering the central area . n = 8 for Cfl1 mutants; n = 13 for controls . A pale grey polyvinyl chloride EPM with the following dimensions was used: arm ( either open or closed ) length: 300 mm , width 50 mm , and height of the closed arms 150 mm . At the beginning of the experiment , mice were placed in a closed arm facing away from the center region . Time spent in open arms , visits to open and closed arms and distance travelled in open arms were assessed using the VideoMot2 video tracking system ( TSE Systems , Bad Homburg , Germany ) . Latency to enter the open arm was recorded manually . Statistical tests were performed using the R statistical package; the Mann-Whitney test was used to compare the percentage time spent in open arms and the number of visits to open arms . n = 8 for Cfl1 mutants; n = 11 for controls . All animal work was conducted according to UK guidelines and approved by the UK Home Office . | Thousands of small effect loci are believed to contribute to behavioural variation in mammals . Their abundance and small size frustrate gene identification and make it difficult to know which among them are central to the responsible biological mechanisms . Using imputed genome sequences from 2 , 000 outbred mice and by testing for an enrichment of functional annotations , we identify 167 candidate genes involved in anxiety . Unexpectedly , annotations implicate actin depolymerizing factors ( ADFs ) , including cofilin-1 ( Cfl1 ) , as being involved with the expression of anxiety phenotypes in mice . We confirmed that forebrain-specific inactivation of Cfl1 decreased anxiety in knockout mice . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"animal",
"genetics",
"functional",
"genomics",
"genetic",
"screens",
"genetics",
"molecular",
"genetics",
"biology",
"genomics",
"genetics",
"and",
"genomics",
"gene",
"function"
] | 2012 | Cofilin-1: A Modulator of Anxiety in Mice |
Discriminating self and non-self is a universal requirement of immune systems . Adaptive immune systems in prokaryotes are centered around repetitive loci called CRISPRs ( clustered regularly interspaced short palindromic repeat ) , into which invader DNA fragments are incorporated . CRISPR transcripts are processed into small RNAs that guide CRISPR-associated ( Cas ) proteins to invading nucleic acids by complementary base pairing . However , to avoid autoimmunity it is essential that these RNA-guides exclusively target invading DNA and not complementary DNA sequences ( i . e . , self-sequences ) located in the host's own CRISPR locus . Previous work on the Type III-A CRISPR system from Staphylococcus epidermidis has demonstrated that a portion of the CRISPR RNA-guide sequence is involved in self versus non-self discrimination . This self-avoidance mechanism relies on sensing base pairing between the RNA-guide and sequences flanking the target DNA . To determine if the RNA-guide participates in self versus non-self discrimination in the Type I-E system from Escherichia coli we altered base pairing potential between the RNA-guide and the flanks of DNA targets . Here we demonstrate that Type I-E systems discriminate self from non-self through a base pairing-independent mechanism that strictly relies on the recognition of four unchangeable PAM sequences . In addition , this work reveals that the first base pair between the guide RNA and the PAM nucleotide immediately flanking the target sequence can be disrupted without affecting the interference phenotype . Remarkably , this indicates that base pairing at this position is not involved in foreign DNA recognition . Results in this paper reveal that the Type I-E mechanism of avoiding self sequences and preventing autoimmunity is fundamentally different from that employed by Type III-A systems . We propose the exclusive targeting of PAM-flanked sequences to be termed a target versus non-target discrimination mechanism .
There are several prokaryotic defense systems that confer innate immunity against invading mobile genetic elements , such as receptor masking , blocking DNA injection , restriction/modification ( R-M ) and abortive infection ( reviewed in [1]–[3] ) . In addition , half of the bacteria , and most of the archaea , contain CRISPR-Cas ( Clustered Regularly Interspaced Short Palindromic Repeats/CRISPR-associated ) defense systems , unique in being the only adaptive line of prokaryotic defense ( reviewed in [4]–[7] ) . CRISPR-Cas systems provide adaptive immunity to the host by incorporating invader DNA sequences into chromosomal CRISPR loci [8]–[11] . The 30–40 nt invader-derived DNA sequences are separated by host-derived similarly-sized repeat sequences . Adjacent to a CRISPR locus , a set of cas genes can often be found that encode the protein machinery essential for CRISPR-immunity . The cas genes occur in characteristic combinations that serve as a classification criterion of CRISPR-Cas systems into three major types [12] . In Type I and Type III systems the long precursor CRISPR RNA ( pre-crRNA ) is processed by CRISPR specific endoribonucleases into small CRISPR RNAs ( crRNAs ) that contain a repeat sequence flaked by portions of the adjacent CRISPR repeat sequence [13]–[18] . In some CRISPR-Cas subtypes the crRNA undergoes further processing at the 3′ end [19] , [20] . In Type II CRISPR-Cas systems the pre-crRNA is processed by RNase III [21] . The processed crRNA molecules then remain bound to one or more Cas proteins to guide recognition and cleavage of complementary nucleic acid sequences [22]–[27] . With the exception of Type III-B CRISPR-Cas systems , which cleave RNA [23] , [24] , [28] , all other characterized CRISPR-Cas systems appear to target DNA [27] , [29]–[32] and hence require a mechanism to avoid aberrant cleavage of genomic DNA , i . e . a mechanism to discriminate the genomic “self” DNA of a CRISPR cassette from the invader “non-self” DNA . The absence of such discrimination leads to a suicidal autoimmune response [33]–[35] . In R-M systems this problem is solved by modification of the genomic DNA and cleavage of unmodified invader DNA only ( reviewed in [3] ) . For CRISPR-Cas systems on the other hand , the mechanism ( s ) of self versus non-self discrimination is only partially understood . For the Type III-A system of Staphylococcus epidermidis autoimmunity is prevented through a mechanism that relies on sensing base pairing between the 5′-handle ( the repeat-derived sequence at the 5′-end of the crRNA ) and the corresponding portion of CRISPR repeat [36] . The Type III-A CRISPR-Cas system consists of nine cas genes ( cas1 , cas2 , cas10 , csm2 , csm3 , csm4 , csm5 , csm6 , cas6 ) and a CRISPR with type-8 repeats [37] . After a primary processing step of the pre-crRNA , the resulting crRNAs are further matured through ruler-based cleavage from the 3′ end , yielding 43 and 37 nt crRNA species [20] . These mature crRNA species guide one or more Cas proteins ( possibly a Csm-complex ) to target DNA [32] , presumably through base pairing between the crRNA spacer sequence and the complementary protospacer sequence . However , CRISPR-interference is inhibited when , in addition to base pairing over the spacer sequence , the 5′-handle also base pairs with the protospacer-flanking sequence of the target DNA [36] . In this manner , self-targeting of the CRISPR locus is avoided by default , since self-targeting inevitably leads to full base pairing of the 5′-handle of the crRNA with the CRISPR repeat sequence from which it is transcribed . In particular , the presence or absence of base pairing at three positions downstream of the protospacer ( positions −2 , −3 , and −4 relative to the 3′-end of the protospacer ) is decisive in discriminating self from non-self [36] . The molecular details of how base pairing at positions downstream of the protospacer are sensed , and whether it involves Cas proteins , is currently unknown . Intriguingly , Type I systems contain di- or tri-nucleotide conserved motifs ( protospacer adjacent motifs ( PAM ) ) downstream of protospacers opposite of the crRNA 5′-handle [38]–[40] ( Figure 1A and 2A ) . In the Type I-E CRISPR-Cas system , PAM sequences are recognized by ribonucleoprotein complex Cascade during target DNA binding [29] , [41] . The Type I-E system of Escherichia coli K12 consists of 8 cas genes ( cas3 , cse1 , cse2 , cas7 , cas5 , cas6e , cas1 , cas2 ) and two CRISPR loci with type-2 repeats [37] . The ribonucleoprotein complex Cascade is composed of a 61 nt crRNA , and five different Cas proteins in an uneven stoichiometry: Cse11Cse22Cas76Cas51Cas6e1 [22] . Cascade efficiently binds target DNA through an R-loop formed between the 32 nt spacer sequence of the crRNA and the protospacer sequence [22] ( Figure 1A ) , with a binding affinity that is strongly dependent on the presence of one of the four functional PAM sequences [29] , [41] . Whereas R-loop formation by Cascade involves the entire protospacer sequence [22] , it is unknown whether the PAM nucleotides can participate in base pairing with the crRNA and , if so , how this influences CRISPR interference . Due to the fact that the last nucleotide from the repeat is derived from the PAM sequence during spacer acquisition [8] , [11] , [42] , this nucleotide in the crRNA invariably has the potential to base pair with the −1 position of the PAM , and therefore might be involved in R-loop formation [8] . In contrast , the −2 and −3 positions of the PAM lack base pairing potential with the 5′-handle of the crRNA ( Figure 2A ) . The 5′-handles of other Type I systems and 3′-handles of Type II also display limited base pairing potential with their cognate PAMs ( Table S1 ) , in principle allowing for a differential base pairing mechanism that defines self versus non-self . For Type I-F CRISPR-Cas systems , potential base pairing between PAM sequences and the 5′-handle of the crRNA was recently shown to affect CRISPR interference [43] , suggesting that self versus non-self discrimination in this subtype may depend both on sensing PAM identity and on sensing differential base pairing with the crRNA repeat . In Type I-E systems it has been shown that a loop structure ( L1 ) of the Cse1 subunit of Cascade specifically interacts with the PAM sequence , a process that is thought to destabilize the double-stranded DNA of the target to allow for strand invasion during R-loop formation [44] . Since self DNA of the CRISPR locus does not contain PAM sequences , this mechanism would specifically direct Cascade to target DNA only . However , the observation that target DNA containing a PAM mutant triggers Cascade-dependent primed spacer acquisition in vivo suggests that PAM authentication may not be absolutely required for R-loop formation [11] . Indeed , negatively supercoiled DNA containing a protospacer with a mutant PAM can still be bound by Cascade , albeit with a lower affinity than the same target with wild-type PAM [29] . In line with this , it was suggested that during phage infection Cascade can overcome the absence of a bona fide PAM when Cascade expression levels are high and that the target flanking sequences could participate in this discrimination event [44] . This suggests that a differential base pairing mechanism may play a role in self versus non-self discrimination by Type I-E CRISPR-Cas systems . In agreement with this , it was suggested that complementarity between the crRNA repeat and the protospacer flanking sequence inhibits CRISPR-interference in the Type I-E system of Streptococcus thermophilus [45] . The mechanistic basis of such a differential base pairing mechanism could lie in a perturbation of Cse1-mediated PAM recognition by base pairing interactions between crRNA repeats and the PAM . To study whether a differential base pairing mechanism plays a role in self versus non-self discrimination by the Type I-E system of E . coli K12 , we have systematically mutated both the crRNA repeats and the protospacer-flanking sequences and determined the effects of these mutations and their combinations on CRISPR interference in vivo and target binding in vitro . The results of our analysis demonstrate that discrimination of self from non-self by Type I-E CRISPR-Cas systems occurs through a mechanism that is independent of base pairing between these sequences . Hence , the principal mechanism by which Type I-E systems discriminate self from non-self appears to be solely Cse1-mediated and as such is fundamentally different from the differential base pairing mechanism employed by Type III-A systems . While the mechanism employed by Type III-A is best described as being based on self-recognition ( self versus non-self ) , the mechanism of Type I-E systems is instead based on target-recognition ( target versus non-target ) . While Type III systems can differentiate between targets and non-targets in the absence of a PAM , Type I-E systems are fully PAM-dependent and discrimination cannot take place in the absence of a PAM .
Self versus non-self discrimination by the Type III-A CRISPR-Cas system of S . epidermidis has been shown to rely on a differential base pairing mechanism [36] . As a result CRISPR-interference is specifically inhibited when protospacer sequences are flanked by CRISPR repeat sequences . To test whether this mechanism also applies to the Type I-E CRISPR-Cas system of E . coli K12 , CRISPR-interference was tested against targets containing protospacers flanked by CRISPR repeat sequences . For these analyses , we have cloned the previously described g8 protospacer , from phage M13 [41] , into the pUC19 plasmid and systematically mutated sequences adjacent to the protospacer . E . coli cells expressing Cascade , a g8 crRNA and Cas3 are resistant against transformation by a plasmid in which the g8 protospacer is flanked by a CAT PAM ( Fig . 1B , pWUR690 , approximately 1000-fold lower efficiency of transformation than a control pUC19 plasmid ) . In contrast , these cells are susceptible to plasmid transformation by plasmid pWUR687 in which the g8 protospacer is flanked by CRISPR repeat sequences ( Figure 1B ) . However , the plasmid resistant phenotype can be restored by introducing a CAT PAM in the CRISPR repeat sequence flanking the protospacer ( pWUR688 ) , which alters the base pairing potential only at the −2 and −3 positions ( Figure 1B ) . Plasmid pWUR689 , which has the potential to base pair with g8 crRNA at positions −1 , −2 and −3 ( protospacer adjacent sequence is CGG ) escapes CRISPR-interference from wild-type g8 crRNA expressing E . coli ( Figure 1B ) . The observation that protospacer adjacent sequences complementary to the crRNA at positions −1 , −2 , and −3 avoid Cascade targeting suggest that base pairing at these positions may play a role in self avoidance . To investigate whether avoidance of targeting is due to decreased binding affinities of Cascade for protospacers with mutations at the −1 , −2 , and −3 positions , we performed Electrophoretic Mobility Shift Assays using purified g8 crRNA-loaded Cascade . While high affinity binding could be demonstrated to dsDNA containing the g8 protospacer flanked by the CAT PAM ( Figure 1B and S1 ) , protospacers flanked by either CRISPR repeat sequences or a repeat-derived CGG sequence were bound with low affinity ( Figure 1B and S1 ) . This indicates that target versus non-target discrimination occurs at the level of Cascade affinity for dsDNA target sequences . Furthermore , the data also indicate that “self” DNA recognition may occur , as observed in Type III-A systems , through sensing differential base pairing between protospacer adjacent sequences and the 5′ handle of the crRNA . To investigate if base pairing between the three nucleotides from the 5′-handle of the crRNA and the PAM is involved in discriminating self from non-self DNA we systematically mutated the corresponding nucleotides in the 5′-handle ( i . e . , −1 , −2 , and −3 ) , and analyzed how these mutations affect CRISPR-based immunity against DNA targets flanked by various PAM sequences . Previously [29] , four PAM sequences ( CAT , CTT , CCT and CTC ) , have been reported to confer immunity on wild-type g8 crRNA expressing E . coli against phage M13 infection in vivo , and to give rise to high affinity DNA binding by g8 crRNA-bound Cascade in vitro ( Figure 2B and Figure S2A ) . The last nucleotide of the 5′-handle of the crRNA ( the −1 position ) invariably has the potential to base pair with the PAM [8] , while the −2 and −3 positions lack such base pairing potential ( Figure 2A ) . The resulting configuration is distinct from the fully base-paired configuration that would form if base pairing in this region were the basis of self versus non-self discrimination . To analyze whether base pairing at position −1 is required for CRISPR interference , a mutant CRISPR was constructed , yielding a g8 crRNA that lacks base pairing potential with the PAM at this position . This CRISPR , denoted g8G-1T carries a G-to-T substitution at position −1 , within the repeat sequence . SDS-PAGE analysis of purified Cascade complexes containing either mutant or WT crRNA shows that these complexes have the same apparent stoichiometry , thereby confirming the integrity of the complex ( Figure S4A ) . In addition , isolation of crRNA from these protein complexes shows that crRNA biogenesis is unaffected by the introduced mutation ( Figure S4B ) . Interestingly , despite the absence of base pairing at the −1 position , cells expressing the mutant crRNA maintain the ability to block infection by M13 phages containing each of the four functional PAM sequences ( Figure 2C ) . Consistently , high affinity binding by g8G-1T crRNA-containing Cascade to targets containing the g8 protospacer and the functional PAM variants was observed ( Figure 2C and Figure S2B ) . However , as previously observed for the WT g8-crRNA-Cascade complex [29] , a mutation at the −2 position of the PAM ( i . e . , CGT ) neither confers resistance in vivo ( efficiency of plaquing ( e . o . p . ) = 1 ) nor gives rise to high affinity DNA binding in vitro ( Figure 2C , and Figure S2B ) . This PAM mutant potentially yields an additional base pair with the −2 position of the 5′-handle , both in the WT g8-crRNA-Cascade and the g8G-1T mutant complex ( Figure 2BC ) . Hence , it appears that a base pair at position −2 may be the signal that a protospacer is located in “self” DNA and therefore should not be targeted . To specifically test the role of base pairing at position −2 in CRISPR-immunity , we designed a synthetic CRISPR locus containing a C to A substitution at the −2 position of a CRISPR locus containing spacer sequences that target the g8 protospacer from M13 phage . The g8C-2A CRISPR mutation results in a slight effect on Cascade assembly , as the bands corresponding to Cse1 and Cse2 have modestly lower and higher intensities on an SDS-PAGE , respectively , as compared to wild-type g8-crRNA-Cascade ( Figure S4 ) . However , g8C-2A CRISPR RNA processing is unaffected ( Figure S4 ) . Importantly , the g8C-2A crRNA-guided Cascade complex has a slightly reduced affinity ( 60±12 nM ) for dsDNA targets that have a canonical CTT PAM sequence , which has the potential to base pair at the −2 position of the mutant crRNA ( Figure 3A , white PAM ) . Despite the potential of the mutant Cascade complex to establish an additional base pair , a partially resistant phenotype ( e . o . p . ∼10−2 ) is observed against phages carrying the canonical PAM ( Figure 3A ) , which is consistent with the in vitro DNA binding experiments ( Figure 3A and Figure S3A ) . Targets containing non-canonical PAM sequences are bound with more reduced affinities by the g8C-2A crRNA-guide Cascade complex and are not subject to CRISPR-interference in vivo ( Figure 3A ) . The partial resistant phenotype of the g8C-2A mutant that is observed in combination with the canonical PAM indicates that potential base pairing at both positions −1 and −2 does not serve as a trigger for a non-targeting response . To probe the importance of base pairing at the −3 position , an additional CRISPR mutant was designed , denoted g8C-3G , which carries a C to G mutation at the −3 position of the CRISPR repeat . Again , complex formation and crRNA biogenesis were unaffected by the mutation ( Figure S4 ) . Although the potential for base pairing with most PAM sequences remains the same , a dramatic decrease in both resistance against M13 phage in vivo and DNA binding by g8C-3G-Cascade in vitro is observed ( Figure 3B and Figure S3B ) . The combined results obtained with the three CRISPR mutants indicate that the repeat sequence itself rather than its base pairing potential with the protospacer flanking sequence affects PAM recognition . In order to have a more complete and unbiased analysis of the effects of adding or removing base pairing potential at positions −1 , −2 and −3 , we constructed 26 different PAM sequences adjacent to the g8 protospacer in the M13 phage genome ( Figure 4A , white text on black background ) . All phages were viable as judged by their ability to infect host bacteria lacking the M13-targeting CRISPR ( data not shown ) . The phages were tested for their ability to infect cells expressing each of the 21 different g8 crRNAs with mutated repeat sequences at positions −1 , −2 and −3 . Northern blot analysis showed that processing of mutant g8 crRNAs was unaffected ( data not shown ) . The results reveal that only a small subset of CRISPR repeat mutants confer full phage resistance , and only in conjunction with the four previously validated functional PAM sequences ( Fig . 4 ) . When resistance was observed , it was independent of crRNA-PAM base pairing patterns , but rather appeared to be constrained by a limited number of allowed nucleotides at the −1 , −2 and −3 positions of the 5′-handle , and a fixed number of PAM sequences . Many 5′-handle mutants show a lack of resistance despite the presence of a bona fide PAM in the target and irrespective of the base pairing pattern ( Figure S5 ) . Efficient CRISPR-interference requires the presence of a cytosine at the −2 position of the crRNA repeat ( Figure 4 ) . Substitution of this position to guanidine or uracil interferes with CRISPR-defense . When this position is mutated to an adenosine , a partially resistant phenotype is observed during phage infection in conjunction with the canonical PAM , which is bound with the highest affinity by Cascade in vitro . Presumably this high affinity binding can compensate for the negative effects on DNA binding caused by mutations at the −2 position of the 5′-handle , leading to a partially phage resistant phenotype . Furthermore , CRISPR-mediated phage resistance requires a cytosine at the −3 position . The most likely explanation for the fact that some repeat mutants are not tolerated is that the Cascade subunits involved in binding the 5′-handle exhibit a level of sequence specificity . Although combinations of fully complementary 5′-handles and protospacer flanking sequences do not lead to phage resistance in vivo , this appears to be base pairing independent ( Figure S5 ) , as restoring the wild-type base pairing pattern by altering protospacer flanking sequences fails to rescue the phage-sensitive phenotype . For example , the g8C-3A , C-2T CRISPR fails to provide resistance either against M13 phage with a fully complementary CAT PAM ( Figure 4B ) or against a CTC PAM mutant phage , which is complementary at the −1 position only ( Figure 4C ) . A similar result is obtained when g8C-3A , C-2A CRISPR expressing cells are infected with CTT or CTC PAM phages ( Figure 4D and E ) , indicating that the repeat sequence itself is affecting CRISPR-interference in these instances . Altogether , these data exclude the possibility that the Type I-E system makes use of a differential base pairing mechanism to inhibit self-targeting . The finding that the specificity of PAM recognition is unaffected by its potential to base pair with the 5′-handle is consistent with Cse1 being the only factor involved in PAM recognition [44] . To rule out the possibility that the specificity of PAM recognition by g8-Cascade variants depends on the expression levels of CRISPR-Cas components , the same analyses were performed with an engineered M13 targeting E . coli strain with cas genes fused to inducible promoters [12] . When repeat mutations were introduced into the genomic CRISPR cassette in this strain , identical results were obtained ( Figure S6 ) , showing that the data described here are expression level independent . Previous studies on the S . thermophilus Type II-A CRISPR1/Cas system have revealed differences in PAM specificity and effectivity in either plasmid or phage interference assays [30] , [45] . To test whether the Type I-E CRISPR/Cas system also displays assay-dependent differences in PAM utilization , we generated plasmids carrying the g8 protospacer ( pG8 ) flanked by any of the 26 PAM mutants tested in the phage assays . Transformation of the pG8 variants into E . coli cells expressing Cascade , a g8 crRNA and Cas3 show that the four PAMs ( CAT , CTT , CCT , and CTC ) that provide interference during phage infection also affect plasmid transformation ( resulting in a more than 1000-fold decrease in efficiency of transformation ( e . o . t . ) ) . Apart from these four PAMs , a non-consensus TTT PAM also yields a full resistance phenotype ( Figure S7; >1000-fold decrease in e . o . t . ) , as has been observed before [8] , while M13 phage carrying this non-consensus TTT PAM sequence escape interference ( Figure 4A ) . In addition , ten non-consensus PAMs give rise to a partial resistance phenotype ( Figure S7; e . o . t . <10−1 for CCA , CAA , GAT , CTG , and AGA PAMs; e . o . t . <10−2 for CTA , GTT , TAT , ATT and TTC PAMs ) , which is in line with previously reported partial resistance in S . thermophilus against transformation with a target plasmids carrying non-consensus PAMs [30] . The data show that PAM authentication during CRISPR-based protection is more promiscuous during plasmid transformation than during phage infection .
CRISPR-Cas systems are the only prokaryotic adaptive immune systems described to date . Although initially thought of as a single system , we now know that these systems are structurally and mechanistically diverse . Here we have investigated whether a differential base pairing mechanism to discriminate self from non-self , as described for the Type III-A system of S . epidermidis , also applies to the Type I-E CRISPR-Cas system of E . coli K12 . By systematically mutating the crRNA repeat sequence and the PAM positions , we demonstrate that this Type I-E system does not utilize the potential for base pairing between the 5′-handle and the protospacer flanking sequences to avoid self targeting . The −1 position of crRNA has recently been shown to be invader-derived and hence invariably has the potential to base pair with cognate DNA , both in E . coli [8] , [11] , [42] and in S . thermophilus [45] , [46] . This discovery suggested that base pairing at the −1 position would be critical for target recognition by Cascade , in the same way that nucleotides in the seed region ( nucleotides +1 to +5 , +7 and +8 ) are essential for target recognition [41] . However , our results clearly show that base pairing at position −1 is not essential for CRISPR-interference . It has recently been suggested that the −1 position of the CRISPR repeat could be considered part of the spacer [42] . However , this does not seem appropriate since this nucleotide does not appear to be involved in base pairing with the invading target sequence . The absence of a base pairing requirement for the −1 position might suggest that this position is not available for base pairing due to structural constraints . The −2 position of the crRNA repeat requires the presence of a cytosine for efficient CRISPR-interference ( Figure 4 ) . When this position is mutated to an adenosine , a partially resistant phenotype is observed during phage infection in conjunction with the canonical PAM . Substitution of the −2 position to a guanidine or uracil renders the CRISPR-interference pathway non-functional . Interestingly , mutation of the −2 position to adenosine causes an apparent structural alteration of the Cascade complex . While most subunits are present in the same apparent stoichiometry in the mutant g8C-2A-Cascade as in the wild-type complex , the Cse1 subunit is underrepresented . This might suggest that Cse1 interacts with the −2 position of the repeat and that interaction with this base is important for efficient incorporation of Cse1 into the complex . Like the −2 position , the −3 position requires a cytosine for CRISPR-mediated phage resistance to be manifested . However , complex formation is unaffected in g8C-3G-Cascade ( Figure S4A ) . The −3 , −2 and −1 positions are among the most conserved bases of type 2 repeats [37] . Although the current resolution of the Cascade structure does not allow us to confidently pinpoint the location of the −2 and −3 bases of the 5′-handle of the crRNA , these bases appear to be part of a 5′ hook-like structure that is primarily cradled by the last subunit of the Cas7 hexamer ( i . e . , Cas76 ) [47] . The arch of the crRNA may position the 5′ terminal nucleotides within bonding distance to residues in loop-1 of Cse1 , which is consistent with the assembly defects reported for L1 mutations [44] . However , the resolution of the current Cascade structure and absence of density for L1 in the X-ray crystal structures of Cse1 prevent confident assignment of these interactions . Higher-resolution structures of the Cascade will be critical for a precise understanding how the crRNA and the Cas proteins are arranged in this complex . In some CRISPR systems PAM sequences play an important role during different stages of CRISPR defense . In the Type I-E system of E . coli , PAM sequences are recognized by Cas1 and/or Cas2 during the selection of pre-spacers for integration into the CRISPR [9] . PAM motifs allow the CRISPR adaptation machinery to correctly orient newly acquired spacers into the CRISPR array [38] , [48]–[50] . Interestingly , in Type I-E systems , the PAM selectivity of the CRISPR-adaptation machinery has co-evolved with that of the CRISPR-interference machinery , as the preference for the CTT PAM is observed both during Cas1/Cas2-dependent spacer integration [9] and during target DNA binding by Cascade [29] . In contrast , the E . coli I-F integration machinery appears to select for a PAM that overlaps but differs from the motif that yields optimal interference levels [43] . In this E . coli I-F subtype the PAM was found to be a GG motif at the −1 and −2 positions relative to the protospacer , while an overlapping , but different , motif ( GG at the −2 and −3 positions ) provided optimal interference levels [43] . The presence of a G at position −2 was both required and sufficient for interference . The I-F subtype of Pectobacterium atrosepticum on the other hand requires a GG motif immediately flanking the protospacer for interference , and mutagenesis of the G at position −1 to a T ( which potentially base pairs with the repeat ) gives rise to an escape phenotype [35] . Recently , a new nomenclature has been proposed that takes into account the differences in motif selectivity during spacer integration and CRISPR-interference [51] . PAMs have been shown to be important for CRISPR interference in various Type I and Type II CRISPR-Cas subtypes ( e . g . Type I-A systems in S . solfataricus [40] , Type I-B in Haloferax volcanii [39] , Type I-E in E . coli [29] , Type I-F in P . aeruginosa [52] , E . coli [43] and P . atrosepticum [35] , as well as in Type II-A and II-B systems of Streptococcus pyogenes and S . thermophilus [27] , [30] , [50] , [53] , [54] ) . Recently published x-ray crystal structures of the Cse1 subunit of Cascade [44] , [55] have provided detailed insights into the molecular mechanism of Cascade-mediated recognition of the PAM . The well-conserved L1 loop of Cse1 was shown to directly interact with the PAM sequence and to enhance target DNA affinity in the presence of a bona fide PAM [44] . As such , the Cse1 subunit plays a crucial role in PAM authentication in Type I-E systems [44] . Our data indicate that PAM authentication occurs without the formation of base pairs between the 5′ handle of the crRNA and the PAM . While Cascade-like complexes appear to be common components of Type I systems , the PAM-authenticating protein , Cse1 , is unique to Type I-E systems . This could mean that other Cascade-like complexes , such as the aCascade ( IA-Cascade ) [25] , IC-Cascade [17] the as yet unidentified ID-Cascade , and the Csy-complex ( IF-Cascade ) [26] may have their own specialized PAM-sensing proteins . It has been hypothesized that the large subunits of Type I systems ( Cas8a1 and Cas8a2 ( Type I-A ) , Cas8b ( Type I-B ) , Cas8c ( Type I-C ) , Cas10d ( Type I-D ) , Cse1 ( Type I-E ) , Csy1 ( Type I-F ) ) are homologous to Cas10 proteins associated with the Type III systems [56] , but these predictions await experimental verification . If these predictions are correct they may suggest that PAM recognition is carried out by the large subunit of other CRISPR-Cas subtypes . Under native-like expression levels , the change in affinity of Cascade for a target resulting from the presence or absence of a PAM sequence appears to be sufficient to serve as a robust mechanism to discriminate non-self target sequences ( i . e . protospacers flanked by a PAM ) from non-target sequences ( i . e . protospacers without PAM ) in vivo [44] . Given the absence of PAM sequences in the CRISPR array , self DNA automatically falls into the non-target category and is not subject to interference . For Type III systems , on the other hand , no PAMs have yet been found , suggesting that these systems lack PAMs [23] , [36] . For Type III-A systems it has been shown that differentiation between self DNA and non-self DNA relies on sensing differential complementarity between the 5′-handle of the crRNA and the protospacer-flanking sequence ( Figure 5A ) [36] . This discrimination mechanism is based on specific recognition of self DNA , and is therefore best described by the term self versus non-self discrimination ( Figure 5A ) . Here we demonstrate that self-avoidance by the Type I-E system does not rely on potential base pairing between crRNA repeats and protospacer flanking sequence . Therefore , Cascade lacks the ability to specifically recognize self and relies on specific target DNA recognition through PAM authentication . We argue that PAM authentication is a “target versus non-target” discrimination mechanism ( Figure 5B ) , which is fundamentally different from the “self versus non-self” discrimination mechanism employed by Type III-A systems . Either mechanism is sufficient to avoid targeting of the CRISPR locus on the host genome . In target versus non-target discrimination , self sequences within the CRISPR locus ( i . e . spacers ) automatically belong to the non-target class , since PAM sequences are absent in the CRISPR repeat . Likewise , in self versus non-self discriminating systems target sequences fall in the non-self class . It appears likely that PAM-sensing CRISPR-Cas systems all make use of target versus non-target discrimination . Unlike Type III systems , discrimination between targets and non-targets by Type I-E systems cannot take place in the absence of a PAM . Both discrimination mechanisms , however , are not mutually exclusive . The Type I-F system of E . coli LF82 has been speculated to utilize both target versus non-target discrimination and self versus non-self discrimination [43] , although this hypothesis awaits experimental verification by testing the effect of crRNA repeat mutagenesis on CRISPR interference . By having both mechanisms in place an additional level of security against self-targeting of the host genome could be warranted . The requirement for a more stringent protection against self-targeting could be related to the constitutive gene expression of the Type I-F in E . coli LF82 [43] , whereas the expression of the Type I-E system of E . coli K12 is repressed under laboratory growth conditions [57] , [58] , [59] . The distinct mechanisms of self versus non-self discrimination of Type III-A and target versus non-target recognition of Type I-E have implications for the route that invaders can take to escape CRISPR-interference . While both systems can be evaded by making point mutations in the protospacer [41] , [60] , only the Type I-E system can be evaded by mutations outside the protospacer , specifically in the region containing the PAM . In contrast , escape from Type III-A interference through mutations outside the protospacer seems rather unlikely , as it would typically require three mutations to establish base pairing between the 5′ handle and the protospacer flank [36] .
E . coli BL21 ( DE3 ) strains were used for Cascade purification . Novablue ( DE3 ) cells supplemented with CRISPR plasmid and plasmids expressing cas genes and engineered K12 strains with cas genes fused to inducible promoters were used for phage sensitivity tests and transformation assays . A description of the plasmids and the strains used in this study can be found in the Supplementary Information ( Table S1 ) . Wildtype M13-Cascade was expressed in E . coli BL21 ( DE3 ) and purified as described before [29] , from pWUR408 , pWUR514 and pWUR615 ( Table S1 ) . g8G-1T-Cascade , g8C-2A-Cascade , g8C-3G-Cascade , were expressed from pWUR408 , pWUR514 and either pWUR680 , pWUR682 , or pWUR684 , respectively ( Table S1 ) . pWUR680 , pWUR682 , and pWUR684 were generated by subcloning a synthetic CRISPR ( Table S3 and Table S4 , Geneart ) into pACYC using EcoNI and Acc65I restriction sites . Although BL21 ( DE3 ) contains genomic CRISPR loci , previous analyses by Mass Spectrometry have demonstrated that these expression and purification conditions yield homogeneous Cascade complexes loaded with crRNA species from the overexpression plasmids , and not from the chromosme [22] . Purified Cascade was separated on a 12% SDS-PAGE as described before [22] , and stained using Coomassie Blue overnight , followed by destaining in Millipore water . Nucleic acids were isolated from purified Cascade complexes using an extraction with phenol∶chloroform∶isoamylalcohol ( 25∶24∶1 ) equilibrated at pH 8 . 0 ( Fluka ) and separated on a 6M urea 15% acrylamide gel , as described in [22] , followed by staining with SybR safe ( Invitrogen ) in a 1∶10000 dilution in TAE for 30 minutes . Electrophoretic Mobility Shift Assays were performed as in [29] , using the PAGE-purified oligonucleotides listed in Table S2 , which were annealed and 5′-labeled with 32P γ-ATP ( PerkinElmer ) using T4 polynucleotide kinase ( Fermentas ) . Determining the Kd of the Cascade target DNA interaction was performed as described in [41] . Briefly , the signals of unbound and bound probe were quantified using Quantity One software ( Bio-Rad ) . The fraction of bound probe was plotted against the total Cascade concentration , and the data fitted by nonlinear regression analysis to the following equation: Fraction bound probe = [Cascade]total/ ( Kd+[Cascade]total ) . Mutations of PAM sequence preceding the g8 protospacer were introduced into the M13 phage genome by QuickChange Site-Directed Mutagenesis Kit ( Stratagene ) as described previously ( [41] ) . Repeat mutant library was generated by QuikChange Site-Directed Mutagenesis Kit ( Stratagene ) according to manufacturer's protocol . The g8 CRISPR cassette plasmid targeting the M13 phage gene 8 ( pWUR477-g8 , described in [41] ) was used as template . Mutations were introduced at positions −3 , −2 , or −1 of the repeat preceding the g8 spacer . Cells sensitivity to wildtype and mutant M13 phages was determined by a spot test method as described [41] or using standard plaquing assay . Efficiency of plaquing was calculated as a ratio of the plaque number formed on a lawn of tested cells to the number of plaques on sensitive ( non-targeting ) cell lawn . K12 strains with cas genes fused to inducible promoters and g8 spacer in CRISPR were transformed with 10 ng of plasmid DNA by electroporation . Transformation efficiency was determined as colony forming units for transformants of targeting strain BW40119 ( Table S1 ) per µg DNA . Plasmids containing the g8 protospacer and PAM mutants were ordered synthetically at Geneart , Germany . | CRISPR loci and their associated genes form a diverse set of adaptive immune systems that are widespread among prokaryotes . In these systems , the CRISPR-associated genes ( cas ) encode for proteins that capture fragments of invading DNA and integrate these sequences between repeat sequences of the host's CRISPR locus . This information is used upon re-infection to degrade invader genomes . Storing invader sequences in host genomes necessitates a mechanism to differentiate between invader sequences on invader genomes and invader sequences on the host genome . CRISPR-Cas of Staphylococcus epidermidis ( Type III-A system ) is inhibited when invader sequences are flanked by repeat sequences , and this prevents targeting of the CRISPR locus on the host genome . Here we demonstrate that Escherichia coli CRISPR-Cas ( Type I-E system ) is not inhibited by repeat sequences . Instead , this system is specifically activated by the presence of bona fide Protospacer Adjacent Motifs ( PAMs ) in the target . PAMs are conserved sequences adjoining invader sequences on the invader genome , and these sequences are never adjacent to invader sequences within host CRISPR loci . PAM recognition is not affected by base pairing potential of the target with the crRNA . As such , the Type I-E system lacks the ability to specifically recognize self DNA . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [] | 2013 | Type I-E CRISPR-Cas Systems Discriminate Target from Non-Target DNA through Base Pairing-Independent PAM Recognition |
Eukaryotic cell growth is coordinated in response to nutrient availability , growth factors , and environmental stimuli , enabling cell–cell interactions that promote survival . The rapamycin-sensitive Tor1 protein kinase , which is conserved from yeasts to humans , participates in a signaling pathway central to cellular nutrient responses . To gain insight into Tor-mediated processes in human fungal pathogens , we have characterized Tor signaling in Candida albicans . Global transcriptional profiling revealed evolutionarily conserved roles for Tor1 in regulating the expression of genes involved in nitrogen starvation responses and ribosome biogenesis . Interestingly , we found that in C . albicans Tor1 plays a novel role in regulating the expression of several cell wall and hyphal specific genes , including adhesins and their transcriptional repressors Nrg1 and Tup1 . In accord with this transcriptional profile , rapamycin induced extensive cellular aggregation in an adhesin-dependent fashion . Moreover , adhesin gene induction and cellular aggregation of rapamycin-treated cells were strongly dependent on the transactivators Bcr1 and Efg1 . These findings support models in which Tor1 negatively controls cellular adhesion by governing the activities of Bcr1 and Efg1 . Taken together , these results provide evidence that Tor1-mediated cellular adhesion might be broadly conserved among eukaryotic organisms .
Coordinated cell–cell adhesion is an essential biological process widely employed by organisms throughout the tree of life . In metazoans , cellular adhesion is important for numerous processes ranging from establishment of body plans and maintenance of differentiated tissues to regulation of cancer progression ( reviewed in [1] , [2] ) . Bacterial and fungal species commonly rely on cellular adhesion during mating and conjugation and for maintenance of multicellular biofilms that function as anchored shields against foreign attack by antimicrobial agents . In essence , cellular adhesion has driven important evolutionary benefits across kingdoms , ranging from the evolution of multicellularity in metazoans to drug resistance in bacteria and fungi . Adhesion plays major roles in virulence-associated traits of several fungal pathogens . In Candida albicans , the most pervasive human fungal pathogen , cellular adhesion is essential for biofilm development . C . albicans biofilms form on both biotic and abiotic surfaces ( such as tissues , plastic prosthesis , dentures and catheters ) [3] and function as reservoirs of infective cells that among immunocompromised individuals can cause deep seated and often fatal mycosis [4] . The typical architecture of a biofilm consists of mixed layers of intertwined yeast and hyphal cells stabilized by adhesive interactions among neighboring cells . These cell–cell adhesive interactions are mediated by a set of cell surface displayed adhesins , including the Als proteins and the cell wall protein Hwp1 [5] , [6] , [7] , [8] . The ALS genes ALS1 and ALS3 , two of eight ALS family members , are required for adherent interactions during biofilm formation in both in-vitro and in-vivo models of catheter biofilm formation and appear to have redundant functions [6] . HWP1 , which codes for a cell surface glycoprotein targeted by mammalian transglutaminase that links Hwp1 to proteins on the mammalian cell surface ( reviewed in [9] ) , surprisingly is also required for cell adhesive interactions during biofilm formation [6] , [7] . Notably , Als1 , Als3 and Hwp1 play complementary roles during biofilm formation suggesting that they might interact to promote adhesion between adjacent cellular surfaces [8] . Adhesin regulation in C . albicans occurs primarily at the transcriptional level . During biofilm formation , expression of ALS1 , ALS3 and HWP1 is regulated by the transcription factor Bcr1 [5] . Additional factors , such as the transcription factors Tec1 , and the repressors Nrg1 and Tup1 , have also been implicated in regulating adhesin expression [10] , [11] , [12] , [13] , [14] , [15] , [16] . Furthermore , both ALS3 and HWP1 are developmentally regulated and exclusively expressed in C . albicans hyphae [17] , [18] . This level of regulation falls under the domain of the cAMP-protein kinase-A signaling pathway that regulates yeast-hypha morphogenesis via the transcription factor Efg1 in response to nutritional and environmental cues [19] , [20] . However , aside from the cAMP signaling pathway , little is known about additional molecular pathways that transduce nutritional signals to the multiple transcriptional regulators governing adhesin expression . Our understanding of signaling networks regulating virulence traits in C . albicans in response to nutritional cues has relied heavily on previous knowledge of homologous pathways in the model yeast Saccharomyces cerevisiae . Nevertheless , despite the high similarity in gene content between both species , several lines of evidence now suggest that rewiring of regulatory networks is an increasingly common paradigm in C . albicans [21] , [22] , [23] . Following these lines of evidence , we sought to assess whether rewiring has also occurred in the nutrient responsive signal transduction pathway centered on the globally conserved protein kinase Tor . Tor protein kinases were first identified in S . cerevisiae as the target of the antifungal and immunosuppressive agent rapamycin , which inhibits Tor function as an FKBP12-rapamycin complex [24] . Two Tor proteins , Tor1 and Tor2 , have been characterized in S . cerevisiae and Schizosaccharomyces pombe , whereas a single Tor homolog is present in C . albicans , Cryptococcus neoformans , Drosophila melanogaster , and humans [24] , [25] , [26] , [27] , [28] , [29] , [30] , [31] . Both of the C . albicans and C . neoformans Tor homologs are inhibited by conserved FKBP12-rapamycin mechanisms [27] , [28] . In eukaryotic organisms ranging from yeast to humans , the Tor signaling pathway functions as a global regulator of cellular growth in response to nutritional cues ( reviewed in [32] ) . In S . cerevisiae , inhibition of Tor signaling by FKBP12-rapamycin triggers autophagy , inhibits translation , represses ribosomal gene expression and induces expression of retrograde response ( RTG ) , nitrogen catabolite regulated ( NCR ) and stress responsive ( STRE ) genes ( reviewed in [33] ) . In this study , we have characterized the transcriptional programs regulated by C . albicans Tor1 and demonstrate an evolutionarily conserved paradigm for Tor1 signaling in regulating transcriptional responses to nutrient starvation in both C . albicans and S . cerevisiae . Interestingly , our analysis has also uncovered a novel role for Tor1 signaling in regulating cell–cell adhesion in C . albicans . Tor1 activity represses adhesin gene expression and inhibition of Tor1 promotes cell–cell adhesion , a process that is central to biofilm development . Furthermore , through genetic and molecular approaches we have identified Tor1 as a constituent of the network involving the transcriptional regulators Bcr1 , Efg1 , Nrg1 , and Tup1 that governs adhesin expression . In summary our results forge a link between the nutrient-responsive Tor signal transduction cascade and regulation of cell–cell adhesion in a major human fungal pathogen and suggest that this function could be conserved in more complex organisms , including metazoans .
The goal of this study was to characterize and define Tor1-dependent transcriptional responses in C . albicans . Tor1 is essential in C . albicans ( our unpublished results ) precluding the use of mutant strains lacking Tor1 . To overcome this limitation , we compared the genome-wide transcriptional profile of wild type yeast cells grown in YPD liquid medium at 30°C exposed or not exposed to sublethal concentrations ( 20 nM ) of the Tor1 specific inhibitor rapamycin . Gene expression analysis indicated that the abundance of over 400 transcripts changed upon rapamycin treatment ( Table S1 ) . Among these , 330 transcripts showed a greater than 2-fold reduction in expression levels . Strikingly , the expression of a large cluster of genes ( ∼120 ) encoding components of the translational machinery was downregulated in response to rapamycin treatment ( Table S1 ) . This includes genes encoding cytoplasmic ribosomal proteins , rRNA processing enzymes , several RNA polymerase III subunits , translation initiation and elongation factors , and tRNA synthetases . In addition , expression of mitochondrial ribosomal protein and amino acid biosynthetic genes was also repressed ( Table S2 ) . These results are consistent with previous reports showing coordinate repression of the ribosome biogenesis ( Ribi ) regulon during rapamycin treatment of S . cerevisiae cells [34] , [35] , [36] , [37] , and indicates that the role of Tor1 in translation regulation is evolutionarily conserved in C . albicans . Exposure to rapamycin also resulted in pronounced induction of nitrogen catabolite repressed ( NCR ) genes such as those encoding the permeases Gap2 , Mep2 , Can1 and Can2 and the transporters Hip1 , Dip5 and Dur3 ( Table S2 ) . Several other NCR genes were concurrently induced , including genes coding for allantoicase ( Dal2 ) , arginase ( Car1 ) , glutamate dehydrogenase ( Gdh2 ) , and the transcriptional regulators Ure2 and Gat1 ( Table S2 ) . Expression of GLN3 , which encodes a well-characterized transactivator of the NCR response , did not show significant changes , which could be due to experimental conditions or posttranscriptional regulation of Gln3 by Tor1 signaling . Nevertheless , the coordinate expression of this cluster of genes strongly implicates Tor1 as a regulator of the NCR response in C . albicans , and mirrors the same pattern of expression found during Tor inhibition in S . cerevisiae cells [35] , [36] , [38] , [39] . In addition to inducing NCR gene expression , rapamycin treatment resulted in increased gene expression of the amino acid starvation transcriptional regulator Gcn4 , along with several carboxypeptidases , aminopeptidases , and oligopeptide transporter genes ( Table S2 ) . This transcriptional pattern indicates that rapamycin treated cells perceive the environment as nutrient limiting . Interestingly , our analysis also revealed a unique role for Tor1 signaling in regulating a set of C . albicans-specific transcripts . Rapamycin treatment resulted in strong induction of classic hyphae-induced genes such as those encoding the adhesins Als1 and Als3 , the cell wall proteins Rbt1 and Sun41 , and the hyphae-specific protein Ece1 ( Table 1 ) . Furthermore , expression of the transcriptional regulators Rfg1 , Czf1 and Tec1 , which control hyphal-specific gene transcription required for filamentous growth , was also upregulated ( Table 1 ) . As a control , we also analyzed global expression patterns upon rapamycin treatment of the rapamycin resistant TOR1-1/TOR1 strain [27] . Upon rapamycin exposure , the transcriptional effects observed in rapamycin treated wild type cells were largely absent and for most genes only a modest induction was observed ( Table S3 ) , indicating that the effects of rapamycin are largely mediated by inhibition of Tor1 . Overall , the transcriptional program elicited by rapamycin treatment of C . albicans cells reflects the role of Tor1 as a central component in a nutrient sensing signaling pathway . This role is further highlighted by the striking conservation of transcriptional programs regulated by Tor1 in C . albicans and S . cerevisiae . Moreover , it suggests a broadly conserved role for the Tor1 pathway in linking nutrient sensing to cellular growth . The transcriptional regulation of hyphae-specific genes by Tor1 prompted us to investigate whether Tor1 is required for filamentous growth of C . albicans . Wild type cells were grown on several hyphae promoting agar media and induction of filamentation was examined in the presence or absence of sublethal concentrations of rapamycin . Rapamycin inhibited filamentation of wild type colonies grown on all media tested: SLAD , alkaline M199 ( pH 8 . 0 ) , and Spider media at 37°C . Rapamycin inhibited radial filament growth from the edges of colonies on all three growth media , as well as the colony wrinkling typically observed on the domes of colonies grown on Spider medium , resulting in smooth appearing colonies ( Figure 1A ) . Rapamycin did not inhibit filamentation in the TOR1-1/TOR1 rapamycin resistant strain , demonstrating that Tor1 inhibition blocks hyphal growth under the conditions tested ( Figure 1A ) . In contrast , sublethal concentrations of rapamycin did not inhibit hyphal growth on YPD agar supplemented with 10% fetal bovine serum ( FBS ) , even though wild type cells were sensitive to the fungicidal effects of higher concentrations of rapamycin under these conditions ( data not shown ) . In conclusion , these results indicate that Tor1 is a central regulator of filamentous growth under nitrogen starvation , alkaline growth , and nutrient starvation conditions . Our results are also consistent with previous reports that showed similar rapamycin-induced inhibition of hyphal growth of C . albicans , S . cerevisiae , and C . neoformans [40] , [41] , [42] . Next , the effects of Tor1 inhibition on germ tube formation in liquid culture were examined . Wild type cells grown in either YPD liquid medium supplemented with 10% FBS or RPMI liquid medium showed robust germ tube formation after growth for one hour at 37°C . Addition of a sublethal concentration of rapamycin did not have any discernible effects on germ tube formation or elongation ( data not shown ) . In Spider liquid culture medium , wild type cells also formed germ tubes in the presence of rapamycin . However , after two hours of incubation , cultures treated with rapamycin displayed a striking clearing of the growth culture ( Figure 1B ) . Upon closer inspection , clearing of the culture was found to result from extensive cellular aggregation and flocculation of both yeast and hyphal cells , which was not observed with TOR1-1/TOR1 rapamycin resistant mutant cells ( Figure 1C ) . Remarkably , cellular aggregation was only observed with rapamycin treated wild type cells grown in Spider liquid media . Rapamycin did not elicit this phenotype in cells grown in alkaline M199 ( pH 8 . 0 ) , Lee's , minimal , SLAD , or YPD liquid media at 37°C , even after 7 hours of incubation ( data not shown ) . Cells grown in modified Spider medium containing glucose or glycerol as a carbon source instead of mannitol ( the carbon source typically added to Spider medium ) continued to aggregate in the presence of rapamycin ( data not shown ) , indicating that the specific carbon source utilized does not appear to contribute to rapamycin-induced cellular aggregation . Based on these results we conclude that Tor1 has contrasting roles in regulating morphogenesis of C . albicans cells . On agar surfaces , Tor1 is a positive regulator of filamentation during growth in nitrogen ( SLAD medium ) and nutrient limited ( Spider ) conditions and in response to alkaline growth conditions . In contrast , Tor1 does not regulate hyphal growth in liquid cultures and under certain conditions ( Spider liquid medium ) Tor1 acts as a negative regulator of cellular adhesion . Interestingly , rapamycin treatment of Candida guilliermondii cells also results in cellular aggregation when grown in Spider medium at 37°C ( Figure S1 ) , indicating that Tor1 control of cellular adhesion might very well be conserved among other fungal species . To elucidate Tor1 roles in cellular aggregation , microarray analysis was performed on cells grown in Spider liquid medium ( 37°C ) in the presence and absence of sublethal concentrations of rapamycin for 90 minutes . This analysis revealed the same pattern of downregulated genes as observed in cells grown in YPD media at 30°C ( data not shown ) . However , the pattern of upregulated genes differed between these two conditions . During growth on Spider medium , rapamycin did not induce expression of permeases and transporters but resulted in a potent upregulation of several genes encoding GPI-anchored cell wall proteins . Among these , the adhesin genes ALS1 , ALS3 , and HWP1 and the cell wall protein-coding gene ECE1 were the most highly expressed ( Table S4 ) . Induction of ALS1 , ALS3 , HWP1 and ECE1 expression upon Tor1 inhibition was confirmed by northern analysis ( Figure 2A ) . Surprisingly , upregulation of ECE1 expression was also observed in a TOR1-1/TOR1 strain , although at reduced levels compared to wild type cells treated with rapamycin . The Tor1-independent expression of ECE1 could be due to inhibition of the C . albicans FKBP12 homolog Rbp1 by rapamycin . Accordingly , ECE1 expression was modestly induced in the rpb1/rbp1 rapamycin resistant strain lacking FKBP12 [27] irrespective of the presence of rapamycin ( Figure 2B ) . The cellular aggregation phenotype of rapamycin-treated cells suggests that expression of adhesin proteins in the cell wall has a phenotypic consequence . However , because rapamycin also downregulated the expression of genes involved in maintaining the translational machinery , we assayed whether induction of ALS3 expression indeed resulted in increased Als3 protein and cell wall localization . In accord with the mRNA expression levels , rapamycin treatment increased Als3 in the cell wall of wild type cells ( Figure 2C ) . In contrast , Als3 was undetectable in both untreated cells or in rapamycin-treated als3/als3 mutant cells ( Figure 2C ) . This analysis also showed that Als3 localized exclusively in hyphal cell walls , even in the presence of rapamycin ( Figure 2C ) . We did not observe Als3 localization in yeast cell walls , suggesting that Tor1 regulation of ALS3 expression or subsequent localization is hyphae-specific on Spider liquid medium at 37°C . The adhesins Als1 , Als3 and Hwp1 have been shown to play central roles in mediating cellular adhesion to a variety of host cell surfaces as well as mediating adhesive interactions during C . albicans biofilm formation ( reviewed in [43] ) . Induction of ALS1 , ALS3 , HWP1 and ECE1 expression upon Tor1 inhibition could therefore account for the robust cell aggregation observed in cells exposed to rapamycin . Homozygous als1/als1 , als3/als3 , hwp1/hwp1 , and ece1/ece1 mutants all underwent cellular aggregation in the presence of rapamycin ( data not shown ) . However , cells from an als1/als1 als3/als3 homozygous double mutant strain failed to aggregate in the presence of rapamycin ( Figure 2D ) . Complementation of this mutant strain with either the wild type ALS1 or ALS3 gene restored cellular aggregation ( Figure 2D ) , providing evidence that these two adhesins function coordinately in mediating cellular aggregation upon Tor1 inhibition . Taken together , our results show that Tor1 negatively regulates the expression of the ALS1 , ALS3 , HWP1 and ECE1 genes and , upon Tor1 inhibition , both Als1 and Als3 are functionally expressed at the cell wall and their combined actions mediate cellular aggregation of C . albicans cells . ALS1 , ALS3 , HWP1 and ECE1 are hyphal specific genes whose expression is governed by a variety of signal transduction networks poised to sense environmental and nutritional cues ( reviewed in [44] ) . In S . cerevisiae , loss of function mutations in Tor signaling effectors alters sensitivity of these mutants to the fungicidal effects of rapamycin . Accordingly , in order to identify transcriptional effectors downstream of Tor1 signaling in C . albicans , we screened homozygous null mutants in the transactivators CPH1 , CPH2 , EFG1 , TEC1 , BCR1 , CZF1 , RIM101 and the repressors NRG1 , TUP1 and RFG1 for those with altered rapamycin sensitivity . This screen identified the nrg1/nrg1 and tup1/tup1 strains as rapamycin hypersensitive ( Figure 3A ) . Complementation of the tup1/tup1 strain with a wild type TUP1 allele restored a wild type level of rapamycin sensitivity ( Figure 3A ) . Similarly ectopic expression of NRG1 from the ACT1 promoter in the nrg1/nrg1 mutant strain restored wild type sensitivity to rapamycin ( Figure 3A ) . The rapamycin hypersensitive phenotype of the nrg1/nrg1 and tup1/tup1 mutant strains suggested that these transcriptional repressors could be downstream effectors of Tor1 signaling . Moreover , both repressors have been shown to negatively regulate ALS1 , ALS3 , HWP1 , and ECE1 gene expression [11] , [12] , [13] , [14] , [15] , [16] . Northern analysis revealed that NRG1 transcript levels increased in wild type cells grown in Spider medium at 37°C ( Figure 3B ) . It is puzzling that mRNA levels of this repressor of hyphae-specific genes would increase during hyphae-inducing conditions . However , our experiments were performed by growing cells overnight in YPD liquid medium at 30°C followed by washing and resuspension of cells in Spider media . Induction of NRG1 expression could result from shifting cells from YPD to Spider or from 30°C to 37°C . Addition of rapamycin resulted in decreased NRG1 mRNA abundance between 30 and 90 minutes of treatment ( Figure 3B ) . When the NRG1 gene was expressed from the ACT1 promoter in an nrg1/nrg1 mutant strain , NRG1 mRNA downregulation was no longer observed upon rapamycin treatment ( Figure 3B ) . This finding indicates that Tor1 regulates NRG1 transcriptionally rather than post-transcriptionally . Unexpectedly , expression of TUP1 also increased upon transfer of wild type cells from YPD to Spider medium at 37°C ( Figure 3C ) . In the presence of rapamycin , induction of TUP1 expression was curtailed ( Figure 3C ) , indicating that Tor1 signaling is required for complete induction of this repressor . Rapamycin downregulation of NRG1 expression and of TUP1 induction correlates with increased ALS1 , ALS3 , HWP1 and ECE1 mRNA abundance upon rapamycin treatment ( Figures 2B , 3B , and 3C ) . We conclude that the NRG1 and TUP1 genes are both transcriptional targets of Tor1 signaling , supporting a model in which Tor1 inhibition results in decreased NRG1 and TUP1 mRNA levels thereby relieving ALS1 , ALS3 , HWP1 and ECE1 from transcriptional repression . In S . cerevisiae , a common mechanism by which Tor1 regulates gene expression is by controlling the intracellular localization of transcription factors ( reviewed in [33] ) . Accordingly , we sought to identify C . albicans transcription factors that might be Tor1 effectors in regulating cell–cell aggregation and adhesin expression . Our prior screen of loss of function mutants of known transcriptional regulators identified the repressors NRG1 and TUP1 as Tor1 downstream targets . However , we reasoned that regulation of adhesin expression and cellular aggregation is not essential for cell viability and mutations in regulators of this process might not necessarily show altered rapamycin sensitivity . This panel of mutants was re-screened for those that failed to induce cellular aggregation when grown in liquid Spider medium at 37°C in the presence of rapamycin . This screen revealed two strains , containing loss of function mutations in the regulators BCR1 and EFG1 , in which cells failed to aggregate upon Tor1 inhibition by rapamycin ( Figure 4A and 4B ) . Complementation of bcr1/bcr1 and efg1/efg1 mutants with the wild type BCR1 and EFG1 genes respectively restored cellular aggregation upon rapamycin addition . We were unable to screen nrg1/nrg1 and tup1/tup1 mutant strains since both strains aggregated even in the absence of rapamycin ( data not shown ) . Expression of adhesins was also defective in both bcr1/bcr1 and efg1/efg1 mutant strains . We measured expression levels of ALS1 , ALS3 , HWP1 , and ECE1 by real-time quantitative PCR in wild type cells , and in bcr1/bcr1 and efg1/efg1 mutant strains , following rapamycin exposure for 90 minutes at 37°C . All four genes were robustly transcriptionally induced in wild type cells treated with rapamycin relative to untreated cells , in agreement with northern analysis ( Figures 4C and 2A ) , and this induction was strongly dependent on both BCR1 and EFG1 ( Figure 4C and 4D ) . We also observed residual levels of ALS1 , ALS3 , HWP1 , and ECE1 expression that were independent of Bcr1 and Efg1 , suggesting some functional redundancy between Bcr1 and Efg1 or that additional Tor1 effectors are involved in this process . Based on genetic data showing a requirement for both Bcr1 and Efg1 in rapamycin induced adhesin expression , we hypothesized that Bcr1 and Efg1 might be downstream effectors of Tor1 signaling . We measured expression levels of BCR1 and EFG1 by real time qPCR since mRNA levels are undetectable by northern analysis under our experimental conditions ( growth in Spider media at 37°C for 30 , 60 and 90 minutes ) . Expression levels of BCR1 and EFG1 were low and remained unchanged during rapamycin treatment ( data not shown ) , suggesting post-transcriptional regulation of Bcr1 and Efg1 activity by Tor1 ( data not shown ) . However , we failed to detect the Bcr1 and Efg1 proteins in strains expressing Bcr1 and Efg1 with C-terminal 3XHA epitope tags . Under these conditions , it appears that the abundance of both Bcr1 and Efg1 is below the detection limit of standard techniques . Nevertheless , our genetic data implicates both Bcr1 and Efg1 as effectors of Tor1 mediated regulation of adhesin expression and cell–cell aggregation .
Exposure of S . cerevisiae cells to rapamycin triggers changes in gene expression that mimic those observed during nutrient starvation , such as transcriptional repression of genes required for ribosome biogenesis and induction of genes involved in nitrogen utilization . These and other observations have led to the model that the Tor pathway regulates growth and proliferation in response to nutrients ( reviewed in [33] ) . Consistent with this model , our results strongly suggest that in C . albicans , Tor1 functions in an analogous fashion since rapamycin treatment elicited a similar nutrient scavenging transcriptional response . This response includes the coordinate downregulation of genes required for ribosome biogenesis , translation initiation and tRNA synthesis ( Table S1 ) , and concomitant induction of genes coding for amino acid transporters and peptidases necessary for nutrient acquisition and protein degradation ( Table S2 ) . Exposure to rapamycin also resulted in robust expression of nitrogen catabolite repressed ( NCR ) genes ( Table S2 ) such as those encoding the high affinity ammonium permease Mep2 and the general amino acid permease Gap2 . Thus , Tor1 control of the NCR response also appears to be functionally conserved . Furthermore , in Schizosaccharomyces pombe , the Tor1 homologue tor2+ is also involved in repressing nitrogen starvation-responsive genes [45] , [46] signifying that this role for Tor1 signaling is broadly conserved among fungal species . Our microarray analysis also revealed a novel role for Tor1 in regulating hyphae-specific gene transcription ( Table 1 ) . Expression of the hyphae- nduced-transcripts ALS1 , RBT1 , ALS3 , ECE1 and SUN41 increased upon Tor1 inhibition , implicating Tor1 in governing the yeast-to-hyphae morphological transition . In accord with this line of evidence , we found that Tor1 is required for C . albicans filamentation on agar surfaces under a variety of hyphae-inducing conditions ( Figure 1A ) . These results are consistent with previous reports documenting rapamycin suppression of filamentous differentiation in S . cerevisiae , C . albicans , and C . neoformans [40] , [41] , [42] . Surprisingly , rapamycin did not inhibit hyphal growth in liquid medium . This interesting and paradoxical effect suggests that Tor1 is required for contact-dependent induction of C . albicans hyphal growth on semi solid surfaces and perhaps during contact with host cells or extracellular matrix . In addition , we find that Tor1 is also required for hyphal growth induction during nitrogen limiting conditions ( Figure 1A ) . Recently , there has been increasing evidence-implicating Tor1 signaling in this developmental transition . In S . cerevisiae , Tor1 mediated gene expression of the high affinity ammonium transporter Mep2 is controlled by the GATA transcriptional regulator Gln3 [35] , [36] , [38] , [39] . Similarly , in C . albicans the Gln3 homologue regulates the expression of MEP2 and is a likely effector of Tor1 signaling . Filamentous growth is blocked in mep2/mep2 and gln3/gln3 mutants under limiting nitrogen conditions , mirroring the inhibition of filamentous growth in rapamycin treated wild type cells grown on SLAD medium ( Figure 1A ) [47] , [48] , [49] . The differential regulation of filamentation by Tor1 is a reflection of the complexity inherent within the Tor1 signaling network and of its role as an integrator of diverse signals and filamentous growth . Another unique finding revealed during our analysis of Tor1 function in C . albicans is an unexpected role in regulating cell–cell adhesion . Tor1 dependent cell–cell adhesion appears to be conditional since it is only observed in liquid Spider media suggesting that it requires a signal present in this medium . This conditional regulation of cell–cell adhesion could constitute a specialized signal transduction pathway ( involving Tor1 ) utilized by C . albicans to regulate cell adherence in unique niches of the mammalian host , or upon perception of nutritional or environmental signals . Nevertheless , our data strongly indicates a role for Tor1 signaling in regulating this process . Tor1 dependent cell adherence could arise from changes in cell wall composition upon Tor1 inhibition resulting in non-specific electrostatic interactions . However , several lines of evidence support a model in which Tor1 regulates cell adherence via transcriptional regulation of cell surface adhesins . First , inhibition of Tor1 results in induced expression of genes encoding the adhesins Als1 , Als3 , Hwp1 and Ece1 ( Table S4 and Figure 1A ) , all of which play complementary roles in promoting cell adhesion in in vivo and in vitro models of biofilm formation [6] , [7] , [8] . In particular , both Als1 and Als3 , which have overlapping functions in cell adherence during biofilm development , are also required for Tor1 mediated cellular aggregation ( Figure 2A and 2D ) [6] , [8] . Thus , Tor1 inhibits cell adhesion by promoting repression of adhesin gene expression . Second , the finding that rapamycin induced cellular aggregation and adhesin expression is strongly dependent on the transcriptions factors Efg1 and Bcr1 ( Figure 4 ) further strengthens our model . Both Efg1 and Bcr1 are known transactivators of ALS1 , ALS3 , HWP1 , and ECE1 gene expression and in the context of biofilm formation , Bcr1 has a well-established role in promoting cellular adhesive interactions [6] , [50] , [51] , [52] , [53] , [54] . Our genetic analysis thus strongly suggests that Tor1 functions to negatively regulate Efg1 and Bcr1 activity thereby preventing induction of adhesin expression . Our studies also implicate the adhesin transcriptional repressors Nrg1 and Tup1 in Tor1 mediated cell adhesion [11]–[16] . Rapamycin treatment of wild type cells resulted in downregulation of NRG1 and TUP1 mRNA expression ( Figure 3B and 3C ) , which could result in relief of transcriptional repression at adhesin gene promoters . This model is consistent with the known hyper-flocculant phenotypes of nrg1/nrg1 and tup1/tup1 mutant strains . Interestingly , this level of regulation appears to be transient since Tor1 inhibition did not result in constitutive hyphal induction in contrast to that observed in nrg1/nrg1 and tup1/tup1 mutant strains . Based on these studies , it is evident that Tor1 governs cellular adhesion by multiple mechanisms . One simple model consolidating our observations is that Tor1 controls NRG1 and TUP1 expression through Efg1 and Bcr1 . Efg1 is a known transcriptional activator of TEC1 gene expression [52] , and Tec1 is necessary for BCR1 expression [5] . However , this model is not consistent with the finding that tec1/tec1 mutant cells continue to aggregate in the presence of rapamycin ( data not shown ) . Furthermore , neither bcr1/bcr1 nor efg1/efg1 strains exhibit constitutive hyphal induction , suggesting that steady state levels of NRG1 and TUP1 mRNAs remain unchanged in these mutant backgrounds and arguing that Tor1 regulation of NRG1 and TUP1 gene expression is independent from Efg1 and Bcr1 . Therefore , our results are consistent with a model that under ample nutrient conditions , Tor1 blocks cellular aggregation by promoting expression of the adhesin transcriptional repressors Nrg1 and Tup1 and by downregulating Bcr1 and Efg1 activity ( Figure 5 ) . Conversely , during nutrient limiting conditions or sensing of an environmental signal or upon rapamycin treatment , Tor1 is inactivated leading to adhesin expression and aggregation of cells , a process seminal for C . albicans niche colonization and biofilm formation . Whether these events reflect a synergistic mechanism for maximal adhesin induction , or are differentially regulated by distinct signals in various host environments , requires further dissection . In summary , our results link the nutrient-responsive Tor signal transduction cascade to regulation of cell–cell adhesion in a major human fungal pathogen . Rapamycin also modulates aggregation of C . guilliermondii cells ( Figure S1 ) , and expression of adhesion molecules in mammalian endothelial cells [55] , opening the possibility that Tor1's regulation of cell–cell adhesion might be broadly conserved among organisms including metazoans .
Strains used in this study are listed in Table S5 . Wild type cells were grown at 30°C or 37°C on YPD ( 2% Bacto Peptone , 1% yeast extract and 2% dextrose ) , SLAD ( 1 . 7 g/L Yeast Nitrogen Base without amino acids and ammonium sulfate ( Fisher scientific ) , supplemented with 100 µM ammonium sulfate and 2% dextrose ) , alkaline M199 ( pH 8 ) ( 9 . 5 g/L Medium 199 with Earle's salts and L-glutamine and without sodium bicarbonate ( Gibco ) , buffered with 100 mM HEPES ) , Spider [56] , Spider without mannitol and supplemented with either 2% glucose or 2% glycerol , YPD supplemented with 10% Fetal Bovine Serum ( Gibco ) , RPMI ( RPMI 1640 liquid media with Glutamine and without sodium bicarbonate ( Gibco ) supplemented with 2% dextrose and buffered with 0 . 165 M MOPS , pH 7 ) , Lee's [57] , and SD media . Strains were grown in either solid media containing 2% agar or in liquid cultures in the presence of rapamycin ( LC Laboratories ) or drug vehicle ( 90% EtOH/10% Tween-20 ) . For microarray analysis , cell cultures were grown in YPD medium at 30°C to an O . D600 = 0 . 5 . Cultures were treated with either 20 nM rapamycin or drug vehicle and incubated for 60 minutes at 30°C . Total RNA was extracted using a RiboPure-Yeast RNA extraction kit ( Ambion Inc ) and corresponding cDNA synthesized using an AffinityScript-Multiple Temperature Reverse Transcriptase kit ( Stratagene ) . cDNA generated from drug vehicle treated cells were used as reference and labeled with Cy3 ( Amersham ) and cDNA synthesized from rapamycin treated cells were labeled with Cy5 ( Amersham ) and used as the experimental sample . Labeled cDNA were hybridized overnight at 42°C to a 70-mer C . albicans AROS V1 . 2 oligo microarray set ( OPERON technologies ) printed at Duke University microarray facility . Hybridized arrays were scanned with an Axon GenePix scanner ( GenePix 400B , Molecular Devices ) and data extracted using GenePix Pro 4 software ( Molecular Devices ) . Normalization and expression analysis was performed using GeneSpring ( Silicon Genetics ) and Excel software . For microarray analysis of strains grown on Spider medium , cell cultures were grown in YPD liquid medium at 30°C to an O . D600 = 0 . 5 . Cells were washed twice and resuspended in Spider liquid medium . Cultures were treated with either 20 nM rapamycin or drug vehicle and shaken for 90 minutes at 37°C . All microarray analyses were performed with 4 independent biological replicates . Probability scores were calculated with GraphPad Software ( http://www . graphpad . com/quickcalcs/pvalue1 . cfm ) using two degrees of freedom and values from a t-test on the ratio of median values , which was scaled by the standard deviation of four independent replicates . Significantly modulated genes were selected from those whose expression was above two fold higher or lower than wild type and with a p-value less than 0 . 05 . Filamentation of C . albicans was induced by growing cells on SLAD , alkaline M199 ( pH 8 ) and Spider agar media containing 15 nM rapamycin or drug vehicle and incubated at 37°C for 7 days . Colonies were photographed at a 4 . 5× magnification . For cellular aggregation assays cells were grown in YPD medium to an O . D600 = 0 . 5 , washed twice and resuspended in Spider media . Cultures were treated with either 20 nM or drug vehicle , incubated at 37°C , and photographed immediately after shaking . Cells aliquots were imaged by DIC microscopy at 40× magnification using a Zeiss Axioskop 2 upright microscope . For adhesin northern analysis , cells were grown in Spider media following the same procedure as used for cellular aggregation assays and cells treated with 20 nM rapamycin and drug vehicle for 90 minutes at 37°C . For NRG1 and TUP1 expression analysis , cells were similarly treated and 20 ml aliquots harvested at 0 , 30 , 60 and 90 minutes . Total RNA was isolated using a RiboPure-Yeast RNA extraction kit ( Ambion Inc ) and 30 µg of total RNA loaded onto a 1% formaldehyde agarose gel . Following transfer , membranes were hybridized to radioactive DNA probes for each specific gene ( see Table S6 for primer sequences used ) . Hybridized probe signal was detected using a phosphoimager and quantified with Image Quant 5 . 2 ( Molecular Dynamics ) software . Localization of Als3 in hyphae of wild type and als3/als3 strains was assayed by indirect immunofluorescence using a rabbit polyclonal antiserum raised against a recombinant Als3 N-terminus fragment that was kindly provided by Dr . Scott Filler [58] . Als3 antibodies were enriched by 3 sequential incubations of antiserum , on ice for one hour , with hyphae generated from 3×109 als3/als3 blastopores grown in RPMI at 37°C for 3 hours . Localization of Als3 was performed by growing wild type and als3/als3 mutant strains using the same procedure as used for cell aggregation assays . Cells were treated with 20 nM rapamycin or drug vehicle for 90 minutes at 37°C . After treatment , cell cultures were fixed with 1% formaldehyde and spotted onto poly-lysine coated multi well slides . Slides were incubated with a 1∶50 dilution of enriched Als3-N antiserum and a 1∶50 dilution of anti-rabbit IgG secondary antibody conjugated to Alexa-594 ( Molecular Probes ) . Cells were visualized with a 40× objective on a Zeiss Axioskop 2 upright microscope equipped with an LP 615 filter . Cells were grown in Spider media as above and treated with 20 nM rapamycin and drug vehicle for 90 minutes at 37°C . Total RNA was extracted and cDNA generated using an AffinityScript QPCR cDNA Synthesis kit ( Stratagene ) . RT-qPCR products were obtained using 0 . 15 µg of cDNA , a Brilliant SYBR Green QPCR Master Mix kit ( Stratagene ) and a ABI-7900 light cycler . Primers used for RT-qPCR are described in [59] . Expression levels were calculated by comparative delta Ct and normalized to TDH3 . Expression values are presented as Ct values of rapamycin treated samples relative to TDH3 normalized Ct values of vehicle treated samples . Information for the following C . albicans genes can found at the Candida Genome Database ( CGD ) Web site ( http://www . candidagenome . org ) : ALS1 ( orf19 . 5741 ) , ALS3 ( orf19 . 1816 ) , HWP1 ( orf19 . 1321 ) , ECE1 ( orf19 . 3374 ) , BCR1 ( orf19 . 723 ) , TEC1 ( orf19 . 5908 ) , NRG1 ( orf19 . 7150 ) , TUP1 ( orf19 . 6109 ) , EFG1 ( orf19 . 610 ) , TOR1 ( orf19 . 2290 ) , GAP2 ( orf19 . 6993 ) , MEP2 ( orf19 . 5672 ) , CAN1 ( orf19 . 97 ) , CAN2 ( orf19 . 111 ) , HIP1 ( orf19 . 4940 ) , DIP5 ( orf19 . 2942 ) , DUR3 ( orf19 . 781 ) , CAR1 ( orf19 . 3934 ) , GDH2 ( orf19 . 2192 ) , URE2 ( orf19 . 155 ) , GAT1 ( orf19 . 1275 ) , GLN3 ( orf19 . 3912 ) , GCN4 ( orf19 . 1358 ) , RBT1 ( orf19 . 1327 ) , SUN41 ( orf19 . 3642 ) , SAP10 ( orf19 . 3839 ) , SAP8 ( orf19 . 242 ) , RFG1 ( orf19 . 2823 ) , CZF1 ( orf19 . 3127 ) , RBP1 ( orf19 . 6452 ) , CPH1 ( orf19 . 4433 ) , CPH2 ( orf19 . 1187 ) . Information for the S . cerevisiae TOR1 gene ( YJR066W ) can found at the Saccharomyces Genome Database ( SGD ) Web site ( http://www . yeastgenome . org ) . Microarray data sets can be found at the Gene Expression Omnibus Web site ( http://www . ncbi . nlm . nih . gov/geo/ ) under the accession number GSE13176 . | Living organisms have an intrinsic ability to coordinate their growth and proliferation in response to nutrient availability . In organisms ranging from yeasts to humans , the Tor1 signaling pathway responds to nutrient-derived signals and orchestrates cell growth . Accordingly , we find that in the human fungal pathogen Candida albicans , Tor1 signaling also functions to promote growth . We also uncovered a novel role for the Tor1 molecular pathway in promoting hyphal growth of C . albicans on semi-solid surfaces and in controlling cell–cell adherence . Gene expression analysis and genetic manipulations implicate the known cell surface adhesins Als1 and Als3 as mediators of Tor1-regulated cellular adhesion . Further genetic analysis identified the transcriptional regulators Bcr1 , Efg1 , Nrg1 , and Tup1 that together with Tor1 compose a regulatory network governing adhesin gene expression and cellular adhesion . Given that the Tor pathway is the target of several small molecule inhibitors including rapamycin , a versatile pharmacological drug used in medicine , there is considerable interest in Tor signaling pathways and their function . Moreover , given the potent fungicidal activity of rapamycin against C . albicans , novel antifungal therapies remain to be developed , which may also include novel antifungal therapies with less immunosuppressive rapamycin analogs . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"genetics",
"and",
"genomics/gene",
"expression",
"evolutionary",
"biology/evolutionary",
"and",
"comparative",
"genetics",
"microbiology/microbial",
"physiology",
"and",
"metabolism",
"cell",
"biology/cell",
"growth",
"and",
"division",
"molecular",
"biology/molecular",
"evolution",
"microbiology/microbial",
"growth",
"and",
"development",
"cell",
"biology/cell",
"adhesion",
"evolutionary",
"biology/morphogenesis",
"and",
"cell",
"biology"
] | 2009 | The Protein Kinase Tor1 Regulates Adhesin Gene Expression in Candida albicans |
A randomized controlled field trial to evaluate the effectiveness of a single oral dose of 30 mg/kg of oxfendazole ( OFZ ) treatment for control of porcine cysticercosis was conducted in 4 rural villages of Angónia district , north-western Mozambique . Two hundred and sixteen piglets aged 4 months were selected and assigned randomly to OFZ treatment or control groups . Fifty-four piglets were treated at 4 months of age ( T1 ) , while another 54 piglets were treated at 9 months of age ( T2 ) and these were matched with 108 control pigs from the same litters and raised under the same conditions . Baseline data were collected on the prevalence of porcine cysticercosis using antigen ELISA ( Ag-ELISA ) , as well as knowledge and practices related to Taenia solium transmission based on questionnaire interviews and observations . All animals were followed and re-tested for porcine cysticercosis by Ag-ELISA at 9 and 12 months of age when the study was terminated . Overall prevalence at baseline was 5 . 1% with no significant difference between groups . At the end of the study , 66 . 7% of the controls were found positive , whereas 21 . 4% of the T1 and 9 . 1% of the T2 pigs were positive , respectively . Incidence rates of porcine cysticercosis were lower in treated pigs as compared to controls . Necropsy of 30 randomly selected animals revealed that viable cysts were present in none ( 0/8 ) of T2 pigs , 12 . 5% ( 1/8 ) of T1 pigs and 42 . 8% ( 6/14 ) of control pigs . There was a significant reduction in the risk of T . solium cysticercosis if pigs were treated with OFZ either at 4 months ( OR = 0 . 14; 95% CI: 0 . 05–0 . 36 ) or at 9 months of age ( OR = 0 . 05; 95% CI: 0 . 02–0 . 16 ) . Strategic treatment of pigs in endemic areas should be further explored as a means to control T . solium cysticercosis/taeniosis .
Taenia solium is the etiologic agent of cysticercosis , an important zoonotic infection involving humans and pigs . The life cycle of this parasite includes pigs as the normal intermediate hosts , harbouring the larval cysts in many parts of the body causing cysticercosis , and humans as definitive hosts , harbouring the adult tapeworm in the intestines causing a condition known as taeniosis . Humans are accidental hosts of cysticerci after ingestion of T . solium eggs from the environment and develop the cysts in their tissues and organs , with the central nervous system ( CNS ) being a common site of cyst location resulting in neurocysticercosis [1] , [2] . Cysticercosis in pigs is endemic in many developing countries of Latin America [3] , [4] , Africa [5] , and Asia [6] , where it causes important economic losses resulting from condemnation of infected pork [7] , [8] . The disease has been declared preventable and potentially eradicable [9] , but in many developing countries it is still a major constraint in pig production mainly due to lack of awareness about its extent , poor socioeconomic conditions and the absence of suitable diagnostic tools and control strategies [10]–[13] . Currently , the diagnosis of porcine cysticercosis in live animals is based on lingual examination that is sensitive only in detecting moderate to heavy infections [14] . Reliable serological tests based on detection of specific antibody and antigen have been developed and proved very useful in confirming diagnosis [15] , [16] . Among them the Ag-ELISA has been reported to have high specificity ( 86 . 7% ) and sensitivity ( 94 . 7% ) , even detecting circulating antigens in pigs harbouring one single T . solium cyst [15] , [17] or detecting circulating antigens as early as two to six weeks after infection [17] . However , the detection of circulating antigens technique is unable to distinguish T . solium from T . hydatigena cysticerci , and where the later parasite is highly prevalent the method may be of limited use [18] . Control measures such as improved animal husbandry practices , efficient meat inspection procedures , and health education about hygiene and sanitation have been of limited impact in developing countries where pigs are mainly raised by poor smallholder farmers and marketing of pork is not controlled [19] . However , control of cysticercosis should be possible by eliminating the infection from either pigs or humans , or both for an extended period . Since the pig constitutes a vital link in the transmission cycle of T . solium [20] , [21] , an effective treatment of infected pigs should interrupt the transmission cycle . Oxfendazole ( methyl 5[6]-phenylsulfinyl-2-benzimidazolecarbamate ) ( OFZ ) , a benzimidazole anthelmintic commonly used in cattle and small ruminants for treatment and control of gastro-intestinal roundworms , lungworms and certain tapeworms , has been shown to kill all viable T . solium cysticerci in muscles but ineffective against brain cysts in infected pigs [22]–[25] . Pigs treated with OFZ were reported to be refractory to re-infection even in the event of ongoing exposure to T . solium eggs [25] . More importantly , carcasses from treated pigs were reported to have a normal appearance suitable for human consumption after 3–6 months depending on intensity of infection [23] , [24] . Surveillance for detection of infected pigs followed by treatment with OFZ could reduce the flow of contaminated pork into the market [22] , [23] . Mass porcine chemotherapy with OFZ could , therefore , also be a useful strategy to control T . solium , by providing health as well as economic benefits for rural poor smallholder communities . However , most studies that have addressed the use of OFZ against porcine cysticercosis thus far have mainly focused on efficacy of the drug as opposed to its effectiveness . Therefore , the present study aimed to evaluate the effectiveness of a single oral dose of 30 mg/kg of OFZ against T . solium cysticercosis in pigs reared in smallholder farming systems in a highly endemic area .
The study was conducted in Angónia district located in north-western Mozambique between latitude 14 . 27°S and 15 . 28°S , and longitude 33 . 59°E and 34 . 38°E . The district is characterized by a humid climate with a rainy season extending from November to mid-March and the dry period from April to October . Four villages were selected randomly from a group of 11 villages within the district with a known high prevalence of porcine cysticercosis [26] . A preliminary visit was made to the selected villages to evaluate villagers' willingness to collaborate in a longitudinal study for control of porcine cysticercosis . A randomized field study with a control group was conducted , between August 2008 and May 2009 , to evaluate the effectiveness of a single oral dose of 30 mg/kg of OFZ treatment against T . solium cysticercosis by comparing the prevalence and incidence of porcine cysticercosis between treatment and control groups . Fifty-four pig litters from same number of households , comprising a total of 216 piglets aged 4 months were selected for the study and followed to the age of 12 months . All piglets were identified and ear-tagged with consecutive numbers and each litter in a household was divided into three groups by randomization . The Group 1 piglets ( n = 54 ) were treated with OFZ at 4 months of age ( T1 ) , Group 2 ( n = 54 ) was treated at 9 months of age ( T2 ) while Group 3 ( n = 108 ) served as non-treated controls ( C ) . At the end of the trial , a total of 30 randomly selected pigs ( 8 from each of the OFZ-treatment groups and 14 from the control group ) were purchased from villagers , slaughtered locally and dissected for assessment of T . solium cysticerci . Blood was collected from all animals in both treatment and control groups in three sampling rounds ( 4 , 9 and 12 months of age ) , and information regarding sampling date , household , village , sex and age was recorded . Blood samples were obtained from the cranial vena cava into plain vacutainers tubes and allowed to clot at 4°C . Serum was obtained by centrifugation , dispensed into 2 ml aliquots , stored in labelled vials and kept at −20°C until use . The animals of the two OFZ-treatment groups were first weighed and later given orally a single dose of 30 mg/kg OFZ ( Oxfen-C , Lot 800869 , Bayer , Isando , South Africa ) as a suspension ( concentration of 9 . 06% ) , while the control group did not receive any treatment . The drug was administered through a dispensing tube attached to a 15 ml drench gun . The pig was firmly restrained and a pig snare was used to stabilize the head . The end of the dispensing tube of the drench gun was passed gently , but firmly , over the back of the tongue to allow the pig to swallow the dispensed suspension and ensure the complete delivery of the drug . The Ag-ELISA was performed as described by Brandt and others [27] and modified by Dorny and others [15] . Briefly , the serum samples were pre-treated using trichloroacetic acid ( TCA ) and used in ELISA at a final dilution of 1/4 . Two monoclonal antibodies ( MoAb ) used in a sandwich ELISA were B158C11A10 ( Lot K , ITM , Antwerp , Belgium ) diluted at 5 µg/ml in carbonate buffer ( 0 . 06M/pH 9 . 6 ) for coating and a biotinylated MoAb B60H8A4 ( Lot 28 , ITM , Antwerp , Belgium ) diluted at 1 . 25 µg/ml in phosphate buffered saline-Tween 20 ( PBS-T20 ) +1% new born calf serum ( NBCS ) as detector antibody . The incubation was carried out at 37°C on a shaker for 30 min for the coating of the first MoAb and for 15 min for all subsequent steps . The substrate solution consisting of ortho phenylenediamine ( OPD ) and H2O2 was added and incubated without shaking at 30°C for 15 min . To stop the reaction , 50 µl of H2SO4 ( 4N ) was added to each well . The plates were read using an ELISA reader at 492 nm . Sera from two known positive pigs ( confirmed at slaughter ) were used as positive control . To determine the cut-off , the optical density ( OD ) of each serum sample was compared with a series of 8 reference negative serum samples at a probability level of 0 . 1% using a modified Student's t-test [28] . Carcass dissection was performed as described by Phiri et al [14] with slight modifications . Briefly , skeletal muscle groups were excised from the left half carcasses together with the complete heart , tongue , head and neck muscles , psoas muscles , diaphragm , lungs , kidneys , liver and brains . Dissection was done in such a way that all fully developed cysts could be revealed ( i . e . each slice was about 0 . 5 cm thick ) . Animals were considered infected if viable cysts were found in the carcass . Data were entered and analysed using STATA version 9 . 1 ( Stata Corporation , College Station , TX , USA ) . Prevalence estimates were calculated for pigs that were bled in each sampling round . Incidence was estimated as the number of new cases occurred per unit of animal time at risk , during a period between two consecutive sampling rounds . Confidence intervals were calculated for prevalence and incidence of porcine cysticercosis in both treatment and control groups . Chi-square test and statistical comparison of rates were used to compare the prevalence and incidence , respectively . Logistic regression models were fitted to examine the role of factors potentially associated to prevalence of porcine cysticercosis . Examined factors included treatment time , sex of the pig , pig husbandry practices and village . The statistical significance level was set at 5% . The study was conducted with ethical approval from the scientific board of the Veterinary Faculty , Eduardo Mondlane University . All animals were handled in strict accordance with good animal practice as defined by the OIE's Terrestrial Animal Health Code for the use of animals in research and education . Study permissions were obtained from the Livestock National Directorate , village leaders and pig owners . Due to high level of illiteracy among villagers , an oral consent was obtained from pig farmers in the presence of a witness , who signed on their behalf . At the end of the trial all animals were treated with OFZ , and pig farmers were informed not to slaughter their animals before 4 weeks after the treatment .
All examined pigs ( n = 216 ) were of the local breed ( Landim ) , mostly males ( 55% ) and the majority ( 93 . 1% ) were deliberately left to roam freely . A total of 570 samples were collected along three sampling rounds from 216 pigs . From these animals , 32 ( 14 . 8% ) were sampled once , 184 ( 85 . 2% ) were sampled twice , and 170 ( 78 . 7% ) were sampled three times . Altogether 46 animals were lost to follow-up , out of which 24 ( 22 . 2% ) were from the control group , 12 ( 22 . 2% ) from T1 group and 10 ( 18 . 5% ) were from T2 group . The main reasons for losses of animals to follow-up were death , sale or refusals to allow sampling due to absence of the head of the household . Overall baseline prevalence at 4 months of age was 5 . 1% ( 95% CI = 2 . 6%–8 . 9% ) and did not differ significantly ( p>0 . 05 ) among comparison groups .
This study in Angónia district has evaluated the effectiveness of OFZ treatment in pigs , as a strategy to control T . solium cysticercosis in a highly endemic area in which pigs are constantly exposed to the parasite . A substantial benefit of treating pigs with OFZ using the single oral dose of 30 mg/kg body weight was clearly demonstrated , since the prevalence and incidence in groups of treated pigs was significantly lower compared to the group of untreated pigs . All pigs that were infected at the time of treatment with OFZ were found negative in the subsequent sampling round ( 5 and 3 months later for T1 and T2 , respectively ) . Interestingly , this result was observed in animals raised under constant exposure to T . solium eggs [26] but were in keeping with previous studies conducted under controlled settings reporting a clear effect of OFZ in killing cysts when given as a single dose of 30 mg/kg [22]–[24] , [29] . A significant number of negative control pigs ( 51/79 ) got infected whereas 1 of the 2 infected pigs treated at 4 months and none of the 18 infected pigs treated at 9 months were found re-infected at the end of the study . The latter result is very convincing of induction of a high level protective immunity whereas the former may raise some speculations . One could argue that the result may be explained by differences in individual immune responsiveness or even exposure , as animals become susceptible to new infections after treatment . Also , it can be speculated that there were false serological results or even cross reactions with T . hydatigena infection , though this is unlikely as none of the 51 pigs found infected in control group had fluctuating positivity and all pigs ( 7/30 ) that were found positive at necropsy had no T . hydatigena cysticerci . Nevertheless , in accordance with our findings , naturally infected pigs treated with OFZ under field conditions were not re-infected for at least 3 months . These results support conclusions from previous studies [23] , [25] and speculations that treated pigs remain immune to re-infection for at least 12 weeks [24] . However , although the effectiveness of OFZ can be regarded as good , the prevalence and incidence of T . solium cysticercosis in T1 pigs increased during the study period . The observed increase after treatment is most likely related to new infections in animals being fully susceptible at the time of treatment . Indeed , nearly all of the T1 pigs ( 51/54 ) were negative by Ag-ELISA at the time of treatment , thus being at risk of infection . On the other hand , the prevalence and incidence of T . solium cysticercosis in T2 pigs reduced significantly from treatment at 9 months to 12 months of age . This strategy , although also involving possible treatment of uninfected pigs , showed better results and the potential to be used as a control strategy for T . solium cysticercosis in our settings considering the lifespan of pigs is usually less than one year . Our findings are in line with an earlier field study that showed a clear effect of mass treatment with OFZ at 30 mg/kg body weight in decreasing the prevalence and incidence of porcine cysticercosis , however that study had utilised an antibody-detection method which measures exposure and not necessarily active infection [30] . The significant decrease in the risk of porcine cysticercosis infection at the age of 12 months observed in this study if pigs were treated with OFZ , either at 4 or 9 months of age , was observed in a context of poor living conditions in the study area , where most pigs were left to roam freely and no other control interventions were implemented . This finding , and the low cost of the drug ( approximately 0 . 018 USD/dose ) , corroborates previous conclusions considering OFZ as a potential effective control tool for porcine cysticercosis in resource-poor endemic areas [22]–[24] , [29] . However , it should be borne in mind that many strategies to control T . solium cysticercosis in developing countries have shown promising results [7] , [13] , [21] , [31] , [32] but none has fully succeeded up to date , mainly due to poor socioeconomic and sanitary conditions [7] , [13] . Control measures targeting pigs alone would be ineffective as they cannot prevent the spread of cysticercosis in humans and pigs [21] , [33] , [34] , thus there is a need for integration with other T . solium control strategies in the long term [34]–[36] . Moreover , inexpensive and reliable diagnostic tests in live animals are needed for monitoring the effect of interventions in endemic countries . The Ag-ELISA used in this study , though considered very sensitive [37] may have drawbacks in monitoring the effectiveness of treatment in pigs as cysticercal antigen levels take some time to disappear from circulation after treatment depending on the intensity of infection [24] . Furthermore , the technique may not detect brain cysts [24] , [25] , though in our study brain cysts were only found in some of the control pigs , and currently it does not allow differentiating T . solium from T . hydatigena cysticerci [18] . The results presented in this study showed that OFZ treatment in the last part of the pigs' fattening period is effective to control porcine cysticercosis but is not a stand-alone approach because in high endemic areas a certain number of animals will inevitably get infected after treatment and before slaughter . Although effective , its strategic use as a control tool in T . solium endemic areas should be further explored , particularly with regard to availability , formulation , regimen of administration , safety and marketing of pigs . Despite these concerns and considering that any strategy to control T . solium by targeting pigs has a potential to provide economic incentives to poor smallholder pig farmers , treatment of pigs with OFZ , if integrated with other control measures such as treatment of human tapeworm carriers , ending open human defecation , education , and community-based inspection and sales restrictions , should be considered an important , cost-effective measure to reduce the transmission of T . solium infections in endemic low-income areas . | Porcine cysticercosis is an infection of pigs caused by the larval stage of Taenia solium , a tapeworm that causes taeniosis in humans . The disease is very common in developing countries where it is a serious public health risk and causes significant economic losses in pig production . Many control strategies in developing countries have been of limited impact mainly due to poor socioeconomic and sanitary conditions . An effective treatment of infected pigs using inexpensive drugs may have potential as a long term control tool . We performed a randomized controlled trial to evaluate the effectiveness of oxfendazole treatment for control of porcine cysticercosis . We evaluated the prevalence and incidence of the disease in groups of pigs treated at 4 and 9 months of age and untreated pigs . We found that the prevalence and incidence of the disease in treated pigs was significantly lower than in untreated pigs . We conclude that treatment of pigs with oxfendazole in the last part of the fattening period is cost-effective in controlling porcine cysticercosis in endemic low-income areas but should be integrated with other control measures . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] | [
"veterinary",
"diseases",
"veterinary",
"epidemiology",
"zoonotic",
"diseases",
"veterinary",
"science"
] | 2012 | Use of Oxfendazole to Control Porcine Cysticercosis in a High-Endemic Area of Mozambique |
Trypanosoma cruzi , the causative agent of Chagas' disease , induces multiple responses in the heart , a critical organ of infection and pathology in the host . Among diverse factors , eicosanoids and the vasoactive peptide endothelin-1 ( ET-1 ) have been implicated in the pathogenesis of chronic chagasic cardiomyopathy . In the present study , we found that T . cruzi infection in mice induces myocardial gene expression of cyclooxygenase-2 ( Cox2 ) and thromboxane synthase ( Tbxas1 ) as well as endothelin-1 ( Edn1 ) and atrial natriuretic peptide ( Nppa ) . T . cruzi infection and ET-1 cooperatively activated the Ca2+/calcineurin ( Cn ) /nuclear factor of activated T cells ( NFAT ) signaling pathway in atrial myocytes , leading to COX-2 protein expression and increased eicosanoid ( prostaglandins E2 and F2α , thromboxane A2 ) release . Moreover , T . cruzi infection of ET-1-stimulated cardiomyocytes resulted in significantly enhanced production of atrial natriuretic peptide ( ANP ) , a prognostic marker for impairment in cardiac function of chagasic patients . Our findings support an important role for the Ca2+/Cn/NFAT cascade in T . cruzi-mediated myocardial production of inflammatory mediators and may help define novel therapeutic targets .
Chagas' disease , caused by the infection with the protozoan parasite Trypanosoma cruzi , constitutes the major cause of infectious heart disease in Latin America . It is estimated that 10 million people are infected with T . cruzi in the Central and South America , 100–120 million are at potential risk of infection and around 50 , 000 new cases occur each year [1] . In humans , an acute phase displays frequently as a non-apparent form with a few or no symptoms . Thereafter , the patients enter into an asymptomatic , indeterminate stage , which lasts throughout life in the majority of infected subjects . The remaining 20–30% of chronically infected individuals develop cardiac or digestive complications , typically years or decades after infection . Chronic cardiomyopathy is the most common and severe manifestation of human Chagas' disease , causing congestive heart failure , arrhythmias and conduction abnormalities , which often lead to stroke and sudden death . This type of dilated cardiomyopathy is associated with chronic inflammation and fibrosis , cardiac hypertrophy and thrombo-embolic events [2] . Compromised microcirculation , caused by T . cruzi infection , involves endothelial alterations , vasospasm , reduced blood flow and focal ischemia [3] . Cardiovascular production of vasoactive mediators has been implicated in the pathogenesis of the vasculopathy seen in chagasic heart disease [4] . Among other vasculitis-promoting factors , T . cruzi infection triggers myocardial overexpression and increased plasma levels of endothelin-1 ( ET-1 ) in mice and chronic chagasic patients , which correlate with heart dysfunction [5] , [6] . A bulk of evidence supports the participation of this vasoactive peptide , produced by myocardial and endothelial cells among others , in Chagas' disease pathogenesis [4] , [5] , [7] , [8]–[10] . ET-1 activity may result in vascular injury , cardiac remodeling and enhanced liberation of inflammatory agents [11] . Endothelin-1 is involved in different signaling pathways that include increase in intracellular calcium levels ( [Ca2+]i ) and ERK1/2 activation leading to expression of cyclin D1 and inflammation-linked genes , all of them contributing to T . cruzi-mediated cardiac pathology [12] , [13] . Moreover , ET-1 has been shown to induce cell hypertrophy in primary cultures of rat cardiomyocytes through a calcineurin ( Cn ) /nuclear factor of activated T cells ( NFAT ) -dependent mechanism [14] , [15] . The NFAT family includes four ‘classical’ members displaying a high degree of homology: NFATc1-4 , each of which is expressed in heart tissue [16] . NFAT exists in a highly phosphorylated form in the cytoplasm , which translocates into the nucleus upon dephosphorylation by the phosphatase Cn in response to increases in [Ca2+]i , where it binds to enhancer elements of downstream genes leading to transcriptional activation [17] . One of the NFAT target genes associated with inflammation is cyclooxygenase-2 ( COX-2 ) , the inducible enzyme that catalyzes the rate-limiting step in prostanoid biosynthesis [18]–[20] . ET-1 is able to stimulate protein expression of COX-2 and prostacyclin release in cardiomyocytes [21] . In addition , experimental murine infection with T . cruzi has been shown to raise the number of cardiac cells positive for COX-1 and COX-2 , as well as the circulating levels of cyclooxygenase metabolites [22] , [23] . Both host- and parasite-derived prostaglandins ( PG ) and thromboxane A2 ( TXA2 ) are key regulators of pathogenesis during T . cruzi infection [24] . Remarkably , ET-1 stimulation of cardiac myocytes also results in NFATc4-dependent up-regulation of hypertrophy response genes such as atrial natriuretic peptide ( ANP ) and B-type natriuretic peptide ( BNP ) [25] , [26] , potential markers of myocardial compromise in Chagas' disease [27] , [28] . Although ET-1 and eicosanoids have been proposed to play a role in Chagas' disease pathogenesis , the link between them has not yet been addressed . Thus , we have examined the regulation of Cox2 expression and activity by the combined effect of ET-1 and T . cruzi infection of cardiomyocytes . Our results show that induction of Cox2 expression by ET-1 plus T . cruzi in HL-1 atrial myocytes requires activation of the Ca2+/Cn/NFAT pathway . NFAT is translocated to the nucleus upon stimulation with the peptide and subsequent infection where it binds to NFAT response elements in the promoter region of Cox2 that are essential for transcriptional induction of the gene . Moreover , trypomastigote infection of ET-1-pre-treated HL-1 cardiomyocytes significantly enhanced production of eicosanoids and ANP by these cells . These findings demonstrate the participation of NFAT in [T . cruzi+ET-1]-mediated induction of genes involved in the pathogenesis of chronic Chagas' heart disease .
This study was carried out in strict accordance with the recommendations of Spanish Legislation and the European Council Directive from the Convention for the Protection of Vertebrate Animals Used for Experimental and Other Scientific Purposes . All mice were maintained under pathogen-free conditions in the animal facility at the Centro de Biología Molecular , Universidad Autónoma de Madrid ( Madrid , Spain ) . The animal protocol was approved by the Comité de Ética de la Investigación de la Universidad Autónoma de Madrid . Animals had free access to food and water and were handled in compliance with European codes of practice . Mice were euthanized in a CO2 chamber , and all efforts were made to minimize suffering . Mouse HL-1 cardiomyocytes were plated onto gelatin/fibronectin pre-coated flasks and cultured in Claycomb medium ( Sigma-Aldrich ) supplemented with 10% fetal calf serum , 100 U/ml penicillin , 100 µg/ml streptomycin and 2 mM L-glutamine as previously described [29] . Primary cardiomyocytes were isolated from BALB/c mice and cultured according to standard protocols [30] . HL-1 and primary cardiomyocytes were seeded in 6- ( 5×105/well ) or 24- ( 1×105/well ) well plates and infected with T . cruzi trypomastigotes ( cell∶parasite ratio 1∶5 ) , Y strain , routinely propagated in Vero cells . In some experiments , cell cultures were starved for 18 h and then treated with recombinant murine interferon-γ ( 25 U/ml IFN-γ , R&D Systems ) , 1 µg/ml lipopolysaccharide ( LPS , Sigma-Aldrich ) or 0 . 3 nM ET-1 ( Sigma-Aldrich ) for 2 h before infection . Endotoxin level in the ET-1 batch was <1 EU/mg , as determined using a Limulus amoebocyte lysate analysis kit ( Whittaker Bioproducts ) . Plates were rinsed to remove free parasites and further incubated in complete medium at 37°C , 5% CO2 for the indicated times . Young adult ( 6- to 8-wk-old ) C57BL/6 mice were purchased from Charles River Laboratories . For infection experiments , 2×103 blood trypomastigotes ( Y strain ) per mouse were inoculated by intraperitoneal injection as described [31] , keeping a group of non-infected mice . Age-matched BALB/c mice were infected in parallel . Parasitemia levels were checked every 2 days by direct inspection and counting parasites in a 5 µl drop of tail vein blood . Weekly during one month post-infection , groups of 3 mice were euthanized in a CO2 chamber , and blood and various tissues were collected . Samples were processed for RNA or histological analysis . Total RNA was extracted from HL-1 cells and mouse heart tissue by using Trizol reagent ( Invitrogen ) according to the manufacturer's instructions . First-strand cDNA was prepared by incubation of 1 µg of total RNA with murine leukemia virus reverse transcriptase and random hexamer oligonucleotides ( Bio-Rad Laboratories ) at 40°C for 45 min . Then , 5 µl of the reaction products was amplified by PCR with 1 . 25 U of Taq DNA polymerase ( Invitrogen ) . PCR amplification consisted of 94°C for 45 s for denaturation , 60°C for 45 s for annealing , and 72°C for 45 s for extension , performed for 30 cycles . The sense and antisense primers used for murine Cox2 were: 5′-tcctcctggaacatggactc-3′ and 5′-gctcggcttccagtattgag-3′ , respectively [32] . Aliquots of 10 µl of the PCR products were electrophoresed in a 1 . 6% agarose gel containing ethidium bromide . Quantitative real-time RT-PCR analysis was performed using the High Capacity cDNA Archive Kit ( Applied Biosystems ) , and amplification of different murine genes ( Cox2 , Cox1 , Tbxas1 , Nppa , Edn1 and ribosomal 18S ) was performed in triplicate with the use of TaqMan MGB probes and the TaqMan Universal PCR Master Mix ( Life Technologies ) on an ABI Prism 7900 HT instrument ( Applied Biosystems ) , as reported previously [31] . Quantification of gene expression was calculated using the comparative threshold cycle ( Ct ) method , normalized to the ribosomal 18S control and efficiency of the RT reaction ( relative quantity , 2−ΔΔCT ) . Cardiac tissues from mice were placed after been cut in two pieces in 10% neutral buffered formalin for at least 4 h at room temperature followed by overnight incubation in 70% ethanol . Samples were them embedded in paraffin ( Tissue Embedding Station Leica EG1160 ) , and 5-µm tissue sections were prepared using a motorized Microtome Leica RM2155 . Samples were deparaffinized and rehydrated using a Tissue Processing Station Leica TP1020 . Slides were stained using the Masson's trichrome staining and mounted permanently in Eukkitt's quick hardening mounting system medium ( Biochemika , Fluka Analytical ) . The sections were analyzed in a Leica DMD 108 microscope ( Leica Microsystems , Germany ) . For immunohistochemical studies , myocardial sections were deparaffinized by routine procedures and analyzed using anti-murine COX-2 rabbit polyclonal antibody ( Abcam ) and biotinylated swine antiserum to rabbit immunoglobulin ( Dako ) , following a procedure previously described [33] . Immunoblotting was carried out as described elsewhere [19] . Cardiac cells were disrupted and solubilized extracts ( 20 µg ) were separated in 6% ( only for analysis of NFAT translocation to the nucleus ) or 10% sodium dodecyl sulfate-polyacrylamide gels , and transferred to nitrocellulose filters . After blocking for 2 h with 5% non-fat dried milk in Tris-buffered saline containing 0 . 1% Tween-20 , the membranes were probed 2 h at 37°C with murine monoclonal antibodies against COX-2 ( diluted 1∶250 in blocking buffer , BD Biosciences ) , α-tubulin ( 1∶1000 , Sigma-Aldrich ) , and with rabbit polyclonal antibodies against NFAT ( c1 to c4 isoforms , 1∶200 , Santa Cruz Biotechnology ) , prostaglandin E synthase-2 ( microsomal , 1∶500 ) , thromboxane synthase ( 1∶500 , Cayman ) and prostaglandin F synthase/AK31C3 ( 1∶2 , 000 , ProSci ) . The filters were washed and incubated with the corresponding secondary antibody linked to horseradish peroxidase at 1∶10 , 000 dilution , and the stained bands were visualized by a chemiluminescent peroxide substrate ( Amersham Pharmacia ) . Cox2 promoter constructs spanning from −1796 ( P2-1900-LUC ) and −170 ( P2-274-LUC ) to +104 bp relative to the transcription start site of the human Cox2 gene and the P2-274-LUC plasmid with binding sites for NFAT , or AP-1 , or both mutated were described [19] . The pSH102CD418 expression vector derives from pBJ5 and encodes an NFATc1 deletion mutant ( 1–418 ) that functions as a dominant negative for all NFAT isoforms [34] . HL-1 cells were transfected by Lipofectamine ( Invitrogen ) as described [19] . Briefly , exponential growing cells ( 2×105/well ) cultured in 24-well plates were incubated for 3 h at 37°C with a mixture of 0 . 5–1 µg of the corresponding reporter plasmid and Lipofectamine-containing Opti-MEM ( Invitrogen ) . The total amount of DNA in each transfection was kept constant by using the empty expression vectors . Complete medium was then added to cells and incubated at 37°C for additional 16 h . Transfected cells were exposed to different stimuli ( 0 . 3 nM ET-1 , or phorbol 12-myristate 13-acetate -PMA- plus A23187 calcium ionophore -Ion- , Sigma-Aldrich ) and/or T . cruzi-infected as indicated . In some experiments , FK506 ( 100 ng/ml , Sandoz Ltd . , Tokyo , Japan ) was added for 1 h . Then , cells were harvested and lysed . Luciferase activity was determined by using a luciferase assay system ( Promega ) with a luminometer Monolight 2010 ( Analytical Luminescence ) . Transfection experiments were performed in triplicate . Data of luciferase activity are presented as fold induction ( observed experimental relative luciferase units ( RLU ) /basal RLU in absence of any stimulus ) . Results were normalized for extract protein concentrations measured with a Bradford assay kit ( Pierce , Thermo Fisher Scientific ) . Agonist-induced changes in [Ca2+]i were detected using the Ca2+-sensitive dye Fura-2/AM as described [35] . Briefly , cell monolayers at 80% confluence were trypsinized , washed and then loaded with 1 µM Fura-2/AM under continuous stirring for 30 min at 37°C . The cells ( 2×106/ml ) were exposed to 0 . 3 nM ET-1 and/or infected with T . cruzi trypomastigotes ( cell∶parasite ratio 1∶5 ) , and placed in an Aminco Bowman Series 2 spectrofluorometer ( Thermo ) . Uninfected cultures were used as controls . At the indicated times , the fluorescence signal of Fura-2 was recorded , with excitation and emission at 340 and 510 nm , respectively . Nuclear extracts were prepared from ET-1-treated and/or T . cruzi-infected HL-1 cells as described [36] with minor modifications . Purity of fractions was proven by analyzing cytoplasmic and nuclear marker proteins including α-tubulin ( cytoplasmic ) , and topoisomerase IIβ and c-jun ( nuclear ) . In brief , 5 µg of nuclear protein was incubated with 1 µg of poly ( dI–dC ) DNA carrier in DNA binding buffer ( 10% ( wt/vol ) polyvinylethanol , 12 . 5% ( vol/vol ) glycerol , 50 mM Tris , pH 8 , 2 . 5 mM dithiothreitol , 2 . 5 mM ethylenediaminetetraacetic acid ) for 30 min at 4°C . Then , 105 counts per minute ( c . p . m . ) ( 108 c . p . m . /µg ) of the 32P-labeled double-stranded oligonucleotide ( 2 µg ) were added , and the reaction was incubated at room temperature for 30 min . A synthetic oligonucleotide containing the NFAT consensus sequence 5′-gggtggggtggggaaagccgaggcgga-3′ ( nucleotides −98 to −73 ) in the rat Cox-2 promoter was used as probe/competitor in EMSAs . For competition experiments , a 50-fold molar excess of unlabeled oligonucleotide was added before the addition of the probe . Supershift assays were performed by incubating nuclear extracts with either normal rabbit IgG or anti-NFATc4 antibody for 15 min at 4°C before the addition of the probe . DNA-protein complexes were resolved by electrophoresis in 4% non-denaturing polyacrylamide gels and were subjected to autoradiography . For eicosanoid measurements , HL-1 cells were maintained for 12 h in culture medium supplemented with 0 . 5% fetal calf serum , then pre-treated or not with 10 µM indomethacin ( Sigma-Aldrich ) or 10 µM NS-398 ( Alexis ) for 1 h , and further stimulated with 0 . 3 nM ET-1 for 2 h . After treatment , cardiomyocytes were infected with T . cruzi trypomastigotes for 24 h . At that time , media supernatants were collected and analysed for PGE2 , PGF2α and TXB2 by ELISA ( Cayman ) according to manufacturer's specifications . In addition , eicosanoid levels were determined by ELISA in the sera from both uninfected and T . cruzi-infected C57BL/6 mice at 21 days of infection . For ANP measurements , 24-h supernatants from ET-1-stimulated and/or T . cruzi-infected HL-1 cells , as well as serum specimens from both uninfected and T . cruzi-infected mice , were analyzed by ELISA ( Kamiya Biomedical ) following the instructions of the supplier . For ET-1 measurements , the sera from uninfected and T . cruzi-infected mice were analyzed by ELISA ( Phoenix Pharmaceuticals ) , according to the manufacturer's guidelines . Statistical analysis was performed by using GraphPad Prism 5 . 0 software . Arithmetics means and standard error of the means ( s . e . m . ) were calculated . Significant differences among groups were made by using the one-way analysis of variance test followed by Tukey's test . A difference between groups of P<0 . 05 was considered significant .
As shown in previous works from our group [30] , [37] , C57BL/6 mice proved susceptible to infection with the Y strain of T . cruzi , albeit less severely than BALB/c mice , and survived acute infection ( Figure 1A , B ) . Intense myocardial parasitism and inflammatory pathology were observed at 21 days of infection , together with enhanced COX-2 expression revealed by immunohistochemistry in both cardiomyocytes and heart-infiltrating leukocytes ( Figure 1C ) . Accordingly , T . cruzi-infected C57BL/6 mice showed an augmented ( up to 100 fold ) expression of myocardial Cox2 mRNA ( Figure 1D ) coincident with the highest parasite burden in the heart and maximum severity of myocarditis [30] . In addition , we detected a parallel increase ( up to 15 fold ) in the expression of the TXS gene ( Tbxas1 ) . However , no effect was observed on the expression of Cox1 mRNA ( data not shown ) . Overall , results similar to those above were found in T . cruzi-infected BALB/c mice . Moreover , mRNA levels of ET-1 ( Edn1 ) and ANP ( Nppa ) , a prognostic marker for impairment in cardiac function of chagasic patients [28] , were up-regulated in heart tissue of infected C57BL/6 mice ( Figure 1D ) . Upon infection , ET-1 increased in the two mouse genetic backgrounds . This enhanced mRNA expression in the heart of infected animals was accompanied by elevated serum levels of both peptides and circulating eicosanoids ( TXB2 and PGF2α ) ( Figure 1E ) . It is important to note that observed values from BALB/c and C57BL/6 animals cannot be directly compared to each other , since data are normalized to non-infected values that can differ between both mouse strains . The observed Cox2 mRNA expression in infected heart could come from infected cardiomyocytes , endothelial cells , fibroblasts and/or infiltrating leukocytes . Hence , we tested whether cardiomyocytes up-regulate Cox2 upon T . cruzi infection in vitro . A strong induction of COX-2 protein expression was observed in neonatal cardiomyocyte primary cultures infected with T . cruzi , comparable to that induced by a well-known pro-inflammatory stimulus as LPS plus IFNγ ( Figure 2A ) . To better examine the molecular regulatory mechanism of gene expression of this inducible enzyme by infection , we used the terminally differentiated murine HL-1 cardiomyocyte cell line infected with T . cruzi . Although some reports have described an impaired inflammatory ability of HL-1 cells to express NO synthase-2 or to activate NF-κB [38] , others find the opposite [39] . Nonetheless , in our hands these cells retain contractile and phenotypic characteristics of the adult cardiomyocytes and they are much better suitable for transfection experiments than immature cardiac myocytes , as it has been described [40] . After 3 h of parasite infection , Cox2 mRNA could not be detected . Similarly , a very weak Cox2 induction was also noted in cardiomyocytes cultured in the presence of 0 . 3 nM ET-1 . However , when ET-1-pre-treated HL-1 cells were infected with T . cruzi trypomastigotes ( [T . cruzi+ET-1] ) , a strong increase in Cox2 mRNA expression was detected ( Figure 2B ) . These findings were confirmed by analysing COX-2 protein ( Figure 2C ) . The above results suggested that the combined effect of T . cruzi infection and ET-1 treatment on Cox2 expression was taking place at the transcriptional level . To confirm this , HL-1 cardiac cells were transfected with a Cox2 promoter/luciferase construct spanning from nucleotide −1796 to +104 bp relative to the human Cox2 gene transcription start site ( P2-1900-Cox-2-LUC ) . As shown in Figure 2D , T . cruzi plus ET-1 ( 0 . 3 nM ) induced a four-fold increment ( P<0 . 05 ) in luciferase activity in transiently transfected cells compared to untreated controls . In contrast , T . cruzi-infected cardiomyocytes and ET-1-stimulated uninfected cells showed very little increase . Interestingly , addition of the Cn inhibitor FK506 ( 100 ng/ml ) significantly attenuated [T . cruzi+ET-1]-mediated induction of Cox2 promoter . To map the Cox2 promoter region responsible for [T . cruzi+ET-1] inducibility , we used several Cox2 promoter deletion/mutation constructs . Deletion up to −170 ( P2-1900 to P2-274 ) of the Cox2 promoter region did not significantly affect [T . cruzi+ET-1] inducibility ( Figure 2E ) . Given the relevance of the region spanning from nucleotides −170 to −46 for the recorded induction of the Cox2 promoter , we next determined the contribution of the known transcription factor sites present in this region [19] to the overall transcriptional regulation of [T . cruzi+ET-1]-dependent Cox2 expression . Transfection experiments showed that mutation of the dNFAT ( P2-274 dNFAT mut ) or pNFAT ( P2-274 pNFAT mut ) sites resulted in a 65 and a 60% loss in the [T . cruzi+ET-1]-induced Cox2 promoter activity , respectively , whereas double mutation of both NFAT ( P2-274 p- and dNFAT mut ) sites drastically reduced this activation . Conversely , mutagenesis of the AP-1-like site ( P2-274 AP-1 mut ) present in this region did not significantly diminish the inducibility of the Cox2 promoter by [T . cruzi+ET-1] . To further confirm the central role of NFAT activation in the transcriptional regulation mediated by T . cruzi in ET-1-stimulated HL-1 cells , we co-transfected a dominant-negative version of NFAT ( dnNFAT ) , previously described to abolish NFAT-dependent promoter activity [19] , together with the P2-274-Cox-2-LUC plasmid . Interestingly , expression of dnNFAT abrogated [T . cruzi+ET-1]-induced transcription of the reporter ( Figure 2E ) , supporting the hypothesis of the involvement of NFAT signaling in the regulation of Cox2 gene expression by the cooperation between ET-1 and T . cruzi infection in cardiomyocytes . T . cruzi trypomastigote invasion of cardiac myocytes triggers a transient [Ca2+]i elevation [41] . Similarly , upon the addition of trypomastigotes to HL-1 cells , we observed a transient [Ca2+]i response associated to a considerable , sustained increase in [Ca2+]i during the invasion process ( Figure 3A ) . Comparable outcome , although with higher [Ca2+]i levels , was obtained in T . cruzi-infected HL-1 cells pre-treated with 0 . 3 nM ET-1 . In HL-1 cells , basal expression of several isoforms of NFAT proteins ( c1 , c3 and c4 ) was detected by immunoblot analysis . Interestingly , stimulation with T . cruzi plus ET-1 induced a remarkable increase in the expression of NFATc4 and to a lesser extent , NFATc1 and NFATc3 ( Figure 3B ) . Moreover , NFATc4 was present in the cytoplasm of untreated cardiac cells , but upon parasite infection of ET-1-stimulated cardiomyocytes , it was translocated into the nucleus . Pre-treatment with FK506 ( 100 ng/ml ) , a Cn inhibitor , prevented this translocation , thereby resulting in an accumulation of cytoplasmic NFATc4 protein ( Figure 3C ) . To a much lesser extent , we also observed NFATc1 and NFATc3 migration to the nucleus ( data not shown ) . Together , the above results indicate the activation of the NFATc4 isoform by [T . cruzi+ET-1] through a Ca2+/Cn signaling process . To analyse NFATc4 binding to the NFAT sequences of the Cox2 promoter , we performed EMSAs with nuclear extracts of atrial HL-1 myocytes ( Figure 3D ) . PMA ( 15 ng/ml ) supplemented with Ion ( 1 µM ) was used as a control stimulus . The NFAT oligonucleotide probe from Cox2 promoter specifically bound nuclear proteins from [T . cruzi+ET-1]- and [PMA+Ion]-treated HL-1 cells , which was efficiently competed with a 50-fold molar excess of cold oligonucleotide ( Cox-2-NFAT ) . These inducible complexes were severely diminished in nuclear extracts from cells stimulated with T . cruzi plus ET-1 in the presence of FK506 . No NFAT binding could be demonstrated in response to ET-1 stimulation in the absence of parasites or T . cruzi infection alone . To determine unambiguously the presence of the NFATc4 protein in the complexes , we performed super shifting with an NFATc4-specific antibody . This antibody clearly displaced the migration of the bound probe , allowing the formation of more retarded complexes likely constituted by DNA/NFAT/antibody ( Figure 3D ) . As the NFATc4-specific antibody completely supershifted the complex , it is indicative that c4 , but no other NFAT isoform , is bound to Cox2 promoter DNA in detectable amount . As a negative control , normal rabbit IgG was used . Taken together , these data suggest the binding of NFATc4 to the corresponding sites within the Cox2 promoter in response to T . cruzi infection of ET-1-pre-treated HL-1 cells . To assess whether [T . cruzi+ET-1]-mediated induction of Cox2 expression was associated with an increase in its enzymatic activity , eicosanoid release by HL-1 cells was measured . Compared to mock-treated cells , stimulation of myocytes with 0 . 3 nM ET-1 , or trypomastigote infection over a 24-h period , or the combination of both , induced a significant production of COX metabolites , mainly TXB2 , the stable metabolite of TXA2 , and prostaglandins E2 ( PGE2 ) and PGF2α . Particularly , a striking increase of TXB2 levels , significantly higher than those obtained with T . cruzi and ET-1 separately , was detected in response to [T . cruzi+ET-1] ( Figure 4A ) . Likewise , induction of the Ca2+/Cn/NFAT/COX-2 pathway and eicosanoid production were also achieved in ET-1-primed HL-1 cells exposed to a parasite lysate preparation , thereby suggesting that cardiac cell invasion by trypomastigotes is not absolutely required to produce the cooperative effect with the peptide ( not shown ) . TXB2 , PGE2 and PGF2α synthesis was drastically reduced in the cells incubated with indomethacin ( 10 µM ) , a non-steroidal anti-inflammatory drug known to inhibit both COX-1 and COX-2 enzymatic activity , or with a COX-2-selective inhibitor ( NS398 , 10 µM ) , indicating the important involvement of COX-2 in eicosanoid production upon ET-1 stimulation and T . cruzi infection of HL-1 cardiomyocytes . Treatment of HL-1 cells with COX inhibitors or Cn antagonist had no significant effect on cardiomyocyte-T . cruzi association and did not affect the capacity of the parasites to transform into amastigotes and multiply intracellularly ( not shown ) . Furthermore , analyses for microsomal prostaglandin E synthase-2 ( mPGES-2 ) , prostaglandin F synthase ( PGFS ) and thromboxane synthase ( TXS ) , enzymes that convert the COX product PGH2 to PGE2 , PGF2α and TXA2 , respectively , revealed that [T . cruzi+ET-1] also induced the expression of TXS and PGFS proteins in atrial HL-1 myocytes ( Figure 4B ) . In addition , stimulation with ET-1 promoted a three-fold increased ( P<0 . 05 ) release of ANP . Compared to that observed in mock-treated controls , T . cruzi also up-regulated ANP levels in the supernatants of 24-h-infected cells , which were significantly augmented by the cooperative action of [T . cruzi+ET-1] ( Figure 4C ) .
Trypanosoma cruzi induces multiple responses in the heart , a critical organ of infection and pathology in the host . We herein demonstrated that Cox2 mRNA and protein are induced in mouse heart tissue during T . cruzi infection correlating with cardiac parasite load and myocarditis . This up-regulation was also associated to induction of TXS and of two markers of heart dysfunction previously implicated in Chagas' disease pathogenesis , such as ET-1 and ANP [7] , [10] , [27] . Up-regulation of Cox2 mRNA and protein in myocardial tissue of infected C57BL/6 mice is consistent with a previous report [22] that revealed increased COX-2 protein expression in the heart of infected BALB/c mice . Moreover , several evidences have suggested a role of cyclooxoygenase-derived eicosanoids in the cardiopathogenesis of Chagas' disease ( revised in [42] , [43] ) . Using adult HL-1 atrial myocytes , we further demonstrated that cooperation between T . cruzi and ET-1 stimulated Cox2 mRNA and protein expression leading to the release of eicosanoids . ET-1 seems to be mainly implicated in the establishment of chagasic cardiomyopathy rather than in the control of infection . Previous studies on T . cruzi-infected ET-1 null mice have highlighted the pathogenic role of cardiac myocyte-derived ET-1 in Chagas' heart disease , but these animals did not display higher parasitemia nor lower survival rate than infected wild-type mice [8] . In chagasic heart dysfunction , locally produced ET-1 acts on cardiac myocytes in both an autocrine and/or paracrine manner and chronically induces muscle injury [5] , [7] . In addition , exposure of neonatal rat ventricular cardiomyocytes to ET-1 has been shown to result in higher COX-2 and prostacyclin formation [21] , [44] . In our study , ET-1 induced a dose-dependent increase ( not shown ) in COX-2 activity and eicosanoid biosynthesis in HL-1 cells subsequently infected with T . cruzi . To mimic the pathological microenvironment characteristic of T . cruzi-mediated cardiomyopathy , a 0 . 3 nM ET-1 concentration , close to that detected in the circulation of infected mice and patients exhibiting cardiac involvement [5] , [6] , was selected for pre-treatment of cardiomyocytes . Trypanosoma cruzi invasion of HL-1 cells increased [Ca2+]i , similar to previous report on primary cardiomyocytes [41] . Furthermore , ET-1 induces Ca2+ release in cardiac myofibers [45] . Alterations in [Ca2+]i regulation are frequently recorded in Chagas' disease . In cardiomyocytes from chagasic patients there is a dysregulation of the diastolic [Ca2+]i , while Ca2+ channel blockers display therapeutic potential against chronic chagasic cardiomyopathy [46] , [47] . It has been largely established the requirement for sustained increases , including Ca2+ oscillation frequency , in [Ca2+]i to mediate Cn activation and the nuclear translocation of NFAT [48] . Few studies so far have addressed the impact of T . cruzi infection on the Cn/NFAT pathway in host cells . NFAT has been identified as an important element in innate immunity to T . cruzi and also involved in parasite immune evasion [49] , [50] . The Ca2+/Cn/NFAT pathway has proven functional in adult mouse heart muscle cells and ET-1 has been shown to activate this signaling route in HL-1 atrial myocytes [51] , [52] . Noticeably , NFAT proteins have been described as key molecules for the regulation of Cox2 gene transcription in many different cell types [19] , [53]–[55] . Our present report constitutes the first demonstration that the cooperative effect of ET-1 and T . cruzi infection transcriptionally controls Cox2 expression through activation of the Cn/NFATc4 signaling cascade in cardiomyocytes . Particularly , the two NFAT binding sites in the Cox2 promoter appear to be critical for the observed induction . Mutation of any of these sites strongly diminished Cox2 transcription raised by T . cruzi infection of ET-1-stimulated cardiomyocytes , and dominant negative NFAT prevented that stimulation . Interestingly , this Cn/NFAT pathway has a pivotal role in pathological cardiac hypertrophy [26] . In this regard , we found that ET-1 plus T . cruzi infection leads to enhanced production of the pro-hypertrophic ANP , a prognostic factor for impairment in cardiac function of chagasic patients [28] . Augmented ANP was previously observed in atrial muscle cells upon ET-1 stimulation [56] and , during T . cruzi infection , ET-1 and ANP seem to be important late factors in myocardial remodeling and hypertrophy [10] , [27] . Increased ANP production is somehow linked to the myocardial regulatory pathway induced by [T . cruzi+ET-1] . Thus , PGE2 and PGF2α are known to promote ANP synthesis and release [57] , [58] , while Ca2+ influx is involved in ET-1-triggered ANP expression [59] . More interestingly , NFATc4 was found to regulate several hypertrophy-associated gene transcription in cardiomyocytes , including ANP [26] , [58] . Taken the data together , it is likely that Ca2+ elevation , induced by [T . cruzi+ET-1] , has led to NFATc4 activation , COX-2 induction and augmented ANP secretion by HL-1 cells . A dual role of cyclooxygenase-derived eicosanoids in the course of Chagas' disease has been postulated ( revised in [42] , [43] ) . Morever , the same COX metabolites that mediate host survival during the acute phase may contribute to the progression of cardiac remodeling and heart damage in the chronic phase [60] . The mechanisms involved in the increased prostanoid production in parasite-infected hosts are not yet fully understood . Our findings indicate that the combined effect of ET-1 priming and T . cruzi infection mimics what likely takes place in the heart during infection , inducing eicosanoid-forming enzyme activity through the Ca2+/Cn/NFAT signaling pathway , and leading to enhanced release of prostanoids by atrial cardiomyocytes . Acutely infected mice display elevated PGF2α plasma levels , whereas PGE2 has been found to favor the development of cardiac fibrosis and functional deficits after infection by T . cruzi [23] , [61] . TXA2 , measured as the stable metabolite TXB2 , is the main eicosanoid produced during chronic infection with T . cruzi and this pro-inflammatory agent could be responsible of several of the pathophysiological features of chagasic cardiomyopathy [23] , [24] . TXA2 may exacerbate cardiomyocyte apoptosis , facilitate cytokine biosynthesis by monocytes , activate endothelial cells , and also promote platelet activation , aggregation and degranulation [62] . It is conceivable that the liberated TXA2 might play a role in a feedback loop for ET-1 expression/response , as efficient regulation of ET-1 by a TXA2 mimetic in rat heart smooth muscle cells has been documented [63] . Moreover , the released PGF2α could further induce COX-2 expression and activity , as occurs in carcinoma cells [64] . Enhanced levels of eicosanoids synthesized by [T . cruzi+ET-1]-activated HL-1 cells were down-regulated by addition of COX-2 inhibitors , indomethacin or NS398 . In this regard , meloxicam or etoricoxib , two specific COX-2 inhibitors , minimized the amount of inflammation and fibrosis in the cardiac tissue of infected mice , whereas delayed treatment with aspirin , which blocks COX-1 and COX-2 indistinctly , improved cardiac dysfunction in a murine model of Chagas' heart disease [22] , [60] . However , the potential benefits of COX inhibition for chronic chagasic patients are still unknown . Even though T . cruzi-derived TXA2 and PGF2α have been associated with pathogenesis [24] , [43] , no consistent evidence of parasite COX-2 and TXAS expression is available so far . As we detected overexpression of myocardial enzymes by using mouse-specific probes/antibodies and dampened eicosanoid production in cardiomyocytes treated with mammalian enzyme-specific inhibitors , our data mostly reflect the contribution of prostanoids secreted by host cells to Chagas' myocarditis . In conclusion , we have demonstrated that eicosanoid-converting enzymes are expressed in the infected heart and also that cardiomyocytes respond to ET-1 and T . cruzi infection by induction of COX-2 through activation of the Ca2+/Cn/NFAT intracellular signaling pathway . The cooperation between T . cruzi and ET-1 also led to overproduction of eicosanoids and the pro-hypertrophic factor ANP . These results support an important role for NFAT in [T . cruzi+ET-1]-dependent induction of key agents of pathogenesis in chronic chagasic cardiomyopathy . Identification of the Ca2+/Cn/NFAT cascade as mediator of cardiovascular pathology in Chagas' disease advances our understanding of host-parasite relationship and may help define novel potential targets for therapeutic interventions to ameliorate or prevent cardiomyopathy during chronic T . cruzi infection . | Chronic cardiomyopathy is the most common and severe manifestation of human Chagas' disease , caused by the protozoan parasite Trypanosoma cruzi . Among diverse inflammation-promoting moieties , eicosanoids and the vasoactive peptide endothelin-1 ( ET-1 ) have been implicated in its pathogenesis . Nevertheless , the link between these two factors has not yet been identified . In the present study , we found that T . cruzi infection induces gene expression of ET-1 and eicosanoid-forming enzymes in the heart of infected mice . We also demonstrated that HL-1 atrial myocytes respond to ET-1 stimulus and T . cruzi infection by induction of cyclooxygenase-2 through activation of the Ca2+/calcineurin/NFAT intracellular signaling pathway . Moreover , the cooperation between T . cruzi and ET-1 leads to overproduction of eicosanoids ( prostaglandins E2 and F2α , thromboxane A2 ) and the pro-hypertrophic atrial natriuretic peptide . Our results support an important role for NFAT in T . cruzi plus ET-1-dependent induction of key agents of pathogenesis in chronic chagasic cardiomyopathy . Identification of the Ca2+/calcineurin/NFAT cascade as mediator of cardiovascular pathology in Chagas' disease advances our understanding of host-parasite interrelationship and may help define novel potential targets for therapeutic interventions to ameliorate or prevent cardiomyopathy during chronic T . cruzi infection . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] | [
"medicine",
"immunology",
"microbiology",
"host-pathogen",
"interaction",
"parasitic",
"diseases",
"parasitology",
"cardiovascular",
"neglected",
"tropical",
"diseases",
"signaling",
"pathways",
"calcium",
"signaling",
"infectious",
"diseases",
"signaling",
"in",
"cellular",
"processes",
"biology",
"pathogenesis",
"arachidonic",
"acid",
"signaling",
"cascades",
"calcium",
"signaling",
"cascade",
"immunopathology",
"signal",
"transduction",
"chagas",
"disease",
"calcium-mediated",
"signal",
"transduction",
"molecular",
"cell",
"biology",
"signaling",
"cascades"
] | 2013 | Trypanosoma cruzi Infection and Endothelin-1 Cooperatively Activate Pathogenic Inflammatory Pathways in Cardiomyocytes |
Yeast DNA polymerase ε ( Pol ε ) is a highly accurate and processive enzyme that participates in nuclear DNA replication of the leading strand template . In addition to a large subunit ( Pol2 ) harboring the polymerase and proofreading exonuclease active sites , Pol ε also has one essential subunit ( Dpb2 ) and two smaller , non-essential subunits ( Dpb3 and Dpb4 ) whose functions are not fully understood . To probe the functions of Dpb3 and Dpb4 , here we investigate the consequences of their absence on the biochemical properties of Pol ε in vitro and on genome stability in vivo . The fidelity of DNA synthesis in vitro by purified Pol2/Dpb2 , i . e . lacking Dpb3 and Dpb4 , is comparable to the four-subunit Pol ε holoenzyme . Nonetheless , deletion of DPB3 and DPB4 elevates spontaneous frameshift and base substitution rates in vivo , to the same extent as the loss of Pol ε proofreading activity in a pol2-4 strain . In contrast to pol2-4 , however , the dpb3Δdpb4Δ does not lead to a synergistic increase of mutation rates with defects in DNA mismatch repair . The increased mutation rate in dpb3Δdpb4Δ strains is partly dependent on REV3 , as well as the proofreading capacity of Pol δ . Finally , biochemical studies demonstrate that the absence of Dpb3 and Dpb4 destabilizes the interaction between Pol ε and the template DNA during processive DNA synthesis and during processive 3′ to 5′exonucleolytic degradation of DNA . Collectively , these data suggest a model wherein Dpb3 and Dpb4 do not directly influence replication fidelity per se , but rather contribute to normal replication fork progression . In their absence , a defective replisome may more frequently leave gaps on the leading strand that are eventually filled by Pol ζ or Pol δ , in a post-replication process that generates errors not corrected by the DNA mismatch repair system .
The accuracy by which DNA polymerases synthesize DNA is essential for maintaining genome stability and preventing carcinogenesis . Eukaryotes utilize many DNA polymerases , with different properties , during DNA replication and in DNA repair [1] . DNA polymerase δ ( Pol δ ) , DNA polymerase ε ( Pol ε ) and DNA polymerase α ( Pol α ) ( with associated primase activity ) are required for bulk synthesis of DNA during chromosomal replication [2] . Several studies have suggested that there is a division of labor between Pol δ and Pol ε at the replication fork . Genetic and biochemical studies position Pol δ on the lagging strand [3]–[6] , whereas Pol ε was shown to participate in the synthesis of the leading strand in S . cerevisiae [7] . These studies were preceded by genetic experiments showing that Pol ε and Pol δ proofread opposite strands [8]–[10] . In addition , the Pol ε 3′→ 5′ –exonuclease activity , contrary to the Pol δ 3′→ 5′ –exonuclease activity , does not participate in the correction of errors made by Pol α . This suggests that the proofreading function of Pol ε is restricted to the leading strand [11] , while the exonuclease activity of Pol δ , or perhaps another exonuclease , may proofread both strands [12] . The organization of the replication fork during normal DNA replication , with Pol ε on the leading strand and Pol δ on the lagging strand [6] , [7] , can be disrupted by DNA lesions or sequence contexts in an undamaged template that influence the ability of the replicative polymerase to remain processive [12]–[14] . When polymerases dissociate , the replication machinery must accommodate to complete the replication process and if possible maintain high fidelity . To accomplish this , a variety of strategies are used , including translesion synthesis and recombination pathways [15] . DNA lesions which disengage Pol δ or Pol ε result in single-stranded gaps which are filled in during post-replication repair [16]–[18] . Furthermore , biochemical experiments have shown that collisions between DNA polymerase and transcribing RNA polymerase leads to the abortion of DNA synthesis followed by a reinitiation event when the RNA transcript is used as a primer [19] . To summarize , post-replication repair processes , uncoupled from the replication fork , are likely to occur on both leading and lagging strands to complete DNA replication . Pol α , Pol δ and Pol ε are all composed of several subunits encoded by separate genes . Besides the catalytic subunit , Pol2 ( 256 kDa ) , yeast Pol ε consists of three auxiliary subunits , Dpb2 ( 79 kDa ) , Dpb3 ( 23 kDa ) and Dpb4 ( 22 kDa ) [20] . DPB2 is an essential gene in yeast with an unknown function [21] , yet it is required for early steps in DNA replication and is regulated by Cdc28 kinase [22] , [23] . Recently dpb2 mutations that increase spontaneous mutagenesis were found in S . cerevisiae , suggesting that the second subunit contributes to the fidelity of DNA replication by an unknown mechanism [24] , [25] . DPB3 and DPB4 are non-essential genes . Deletion of DPB3 was previously shown to result in a modest mutator effect [26] , [27] . Dpb3 and Dpb4 both contain histone fold motifs that are known to be important in protein-protein interactions [28] , [29] . Interestingly , Dpb4 is a component of a chromatin-remodeling complex in S . cerevisiae , ISW2 , corresponding to the CHRAC complex found in Drosophila and humans [30] , [31] . The structure of the Pol ε holoenzyme revealed two large domains separated by a flexible hinge [32] . It was suggested that the tail domain of Pol ε was comprised of the Dpb2 , Dpb3 and Dpb4 subunits and was important for the binding to and association with the primer-template dsDNA during DNA synthesis [32] . A purified Dpb3-Dpb4 heterodimer was shown to possess dsDNA binding properties , which in part could explain the properties of the tail-domain [29] . However , this does not exclude the possibility that Dpb2 by itself has properties which allow the tail-domain to interact with dsDNA even without Dpb3 and Dpb4 . In this work , we address whether the Dpb3 and Dpb4 subunits have an effect on the biochemical properties of Pol ε and the fidelity of replication in yeast via a function at the tail-domain of Pol ε . We find that Dpb3 and Dpb4 are important for the processivity of Pol ε polymerase and exonuclease activities , suggesting a role of these two subunits in stabilization of Pol ε interaction with primer-template DNA . Evidently this indirectly affects the fidelity of the overall DNA replication process , since deletion of DPB3 and DPB4 increases both spontaneous frameshift and base substitution mutagenesis , despite an unchanged fidelity of the purified Pol2/Dpb2 complex . A genetic analysis suggests that REV3 contributes to the increased mutation rate in dpb3Δdpb4Δ and the mutational intermediates escape correction by the mismatch repair system .
To investigate the in vivo role of the Pol ε accessory subunits Dpb3 and Dpb4 , we constructed yeast strains wherein either DPB3 , DPB4 or both of these genes were deleted . The frequency of spontaneous mutations in these strains was measured in two reversion assays and one forward mutation assay . We studied the his7-2 and lys2::insE-A14 reversion alleles to score frameshift mutations . The his7-2 allele contains a single base pair deletion in a run of 8 T ( A ) and revert via +1 insertions or -2 deletions [33] . The lys2::insE-A14 allele contains a homonucleotide run of 14 T ( A ) and revert mainly via -1 mutations [34] . The forward mutation assay scores various types of mutations that inactivate the CAN1 gene and result in resistance to canavanine . We found that the dpb3Δ dpb4Δ double deletion has a moderate mutator effect in all assays . Mutation rates for his7-2 reversions and lys2::insE-A14 were increased 2 . 7 and 2 . 6-fold when compared to the wt E134 strain ( Table 1 ) . The mutation rate in the forward mutation assay for canavanine resistance was increased 7 . 4-fold compared to the wt strain ( Table 1 ) . The individual contribution of dpb3Δ or dpb4Δ was comparable to the effect of the deletion of both these genes ( dpb3Δdpb4Δ ) ( Table 1 ) . A proofreading deficient allele of the catalytic subunit , pol2-4 , introduced in the same genetic background resulted in an elevation of the mutation rates similar to the dpb3Δdpb4Δ strain ( Table 1 ) . To determine if the participation of Pol ε in DNA replication depends on DPB3 and DPB4 , we combined dpb3Δ , dpb4Δ , or dpb3Δ dpb4Δ with the pol2-4 mutation . The analysis revealed different genetic interactions . Combining dpb3Δ and pol2-4 led to an additive effect on his7-2 reversion ( Table 1 ) . A higher than additive increase in mutation rate was observed with the his7-2 allele when pol2-4 was combined with dpb4Δ or dpb3Δ dpb4Δ ( Table 1 ) . Reversions scored in the lys2::insE-A14 allele revealed a close to epistatic interaction between pol2-4 , dpb3Δ , dpb4Δ , and dpb3Δ dpb4Δ ( Table 1 ) . The pol2-4 mutation itself elevated the reversion rate of the lys2::insE-A14 allele 2 . 7-fold , which agrees with previous results [35] , [36] . The forward mutation assay with the CAN1 gene revealed an additive effect of the pol2-4 mutation and the double dpb3Δ dpb4Δ deletion . An additive interaction was also found in the pol2-4dpb3Δ strain , but the combination of pol2-4 and dpb4Δ gave a higher than additive increase in mutation frequency ( Table 1 ) . The disparate genetic interactions of DPB3 and DPB4 with the proofreading activity of Pol2 could be due to the separate function of Dpb4 in a chromatin remodeling complex , ISW2 [30] , [31] , [37] , [38] . However , there are no reports demonstrating that ISW2 influence the mutation rate in S . cerevisiae . Another possibility could be that Dpb3 and Dpb4 influence the fidelity of DNA synthesis by Pol ε . To measure the fidelity of Pol ε lacking Dpb3/Dpb4 , we purified the wild type ( i . e . , exonuclease proficient ) Pol2/Dpb2 complex and the exonuclease deficient pol2-4/Dpb2 complex , and then measured their fidelity in an M13mp2 gap-filling assay [39] . The lacZ mutant frequency of the DNA synthesis reaction products generated by the wild type Pol2/Dpb2 complex was 0 . 0018 , comparable to the previously reported value of 0 . 0019 for the four-subunit Pol ε [40] . Both values are near the background lacZ mutant frequency of uncopied DNA , indicating that the exonuclease proficient Pol2/Dpb2 complex is highly accurate . The pol2-4/Dpb2 complex was less accurate , as expected because it is proofreading deficient . However , it was no less accurate than the exonuclease-deficient 4-subunit holoenzyme , as indicated by the similar lacZ mutant frequencies observed for both complexes ( Table 2 ) . To analyze if the error specificity of the 2-subunit enzyme differed from that of the holoenzyme , we sequenced 277 independent mutants generated by pol2-4/Dpb2 , and compared the results to those reported in an earlier study [40] of 285 lacZ mutants generated by the holoenzyme . Comparable error specificity was observed ( Table 2 ) for substitutions , frameshifts and other mutations . We conclude that the increased mutation rates in the dpb3Δ dpb4Δ strain is unlikely to be due to a lower fidelity of DNA synthesis by Pol ε per se . The REV3 gene encodes the catalytic subunit of DNA polymerase ζ ( Pol ζ ) , which is known to be a major contributor to both spontaneous and DNA damage inducible mutagenesis in wild type strains and in strains with defects in other DNA polymerases [27] , [41] , [42] . Yeast Pol ζ has relatively high fidelity for single-base insertions and deletions , and somewhat lower fidelity for base substitutions [43] . Deletion of the REV3 gene suppresses mutagenesis in CAN1 in the dpb3Δ dpb4Δ strain but not mutagenesis in the his7-2 or lys2::insE-A14 allele ( Table 3 ) . Thus , the increase in frameshift mutagenesis observed in the his7-2 and lys2::insE-A14 alleles is Pol ζ-independent . The independence of frameshift his7-2 and lys2::insE-A14 reversion from Pol ζ is consistent with an earlier observation that replication defects ( e . g . in Pol δ mutant , the pol3-Y708allele ) cause Pol ζ dependent mutagenesis for base substitutions only [44] . Published results suggest that Pol δ can proofread errors made by Pol α [11] . To ask if Pol δ proofreads errors generated in the dpb3Δ dpb4Δ strain , we combined the proofreading deficient Pol δ allele pol3-5DV with dpb3Δ dpb4Δ . The pol3-5DV dpb3Δ dpb4Δ strain was viable , in contrast to pol3-01 pol2-4 and pol3-5DV pol2-4 haploid strains [9] . The mutation rates in pol3-5DV dpb3Δ dpb4Δ and pol3-5DV in the CAN1 gene were similar ( Table 1 ) . In contrast , the reversion rate of the his7-2 allele was greater than additive in the pol3-5DV dpb3Δ dpb4Δ strain , when compared to the pol3-5DV strain and dpb3Δ dpb4Δ strain . We conclude that Pol δ has the capacity to proofread a fraction of frameshift errors that occur in the dpb3Δ dpb4Δ strain , but there could also be some other 3′→5′ exonuclease that participates in the process . The mismatch repair protein Msh6 is involved in recognizing a subset of replication errors , specifically single base mismatches and small insertion-deletion intermediates [45] . Although less severe than msh2Δ , pms1Δ or mlh1Δ , inactivation of the MSH6 gene results in a strong increase in mutagenesis ( Table 4 ) . For instance , msh6Δ leads to a dramatic increase of lys2::insE-A14 allele reversion rates as a result of single nucleotide deletions ( Table 4 , [34] , [35] ) . To ask whether DPB3 and DPB4 interact with the mismatch repair system we measured the mutation rates in strains with dpb3Δ and dpb4Δ deletions in an Msh6-deficient background to score for base-base mismatches and small insertion-deletion errors . The combination of the msh6Δ with the dpb3Δ and dpb4Δ gave an additive increase in the his7-2 reversion and can1 mutation rates ( Table 1 ) . The strong synergetic interaction between proofreading defects ( pol2-4 , pol3-01 ) and defects in the mismatch repair system was previously observed for short homopolymeric runs and base substitutions , but not for long homopolymeric runs , such as the A14 run in the lys2::insE-A14 allele [34] , [35] , [46] . In agreement with that , we observed a synergetic interaction between pol2-4 and msh6Δ when mutation rates in a pol2-4 msh6Δ strain were estimated in the his7-2 and CAN1 loci ( Table 1 ) . In the short 8A run of the his7-2 allele , neither the deletion of DPB3 or DPB4 nor both genes affected the mutation rate of the pol2-4 msh6Δ . The multiplicative interaction of pol2-4 and dpb4Δ is absent in the msh6Δ background . The mutation rate in the lys2::insE-A14 gave a complex interaction between msh6Δ and dpb3Δ and dpb4Δ . The mutation rate was somewhat lower ( though not statistically significant , see overlapping confidence limits in Table 1 ) when either dpb3Δ or dpb4Δ was combined with msh6Δ , than when the dpb3Δ dpb4Δ was combined with msh6Δ . When pol2-4 was added to the msh6Δ strain , the combination with dpb3Δ , dpb4Δ or dpb3Δ dpb4Δ gave a mutation rate that was one third of the mutation rate in the pol2-4 msh6Δ strain . At present , it is not clear why this small reduction in mutation rate occurs . The lack of a synergetic interaction between dpb3Δ , dpb4Δ and msh6Δ was unexpected and led us to ask if this was also true for other genes that are required for mismatch repair . Msh2 forms a heterodimer with either Msh3 or Msh6 . Thus , msh6Δ strains still have active Msh2-Msh3 which corrects most replication errors . To completely abolish mismatch repair we deleted MSH2 , MLH1 , or PMS1 . The combination of mlh1Δ or pms1Δ with dpb3Δ dpb4Δ did not reveal a strong synergetic interaction on the his7-2 reversion or can1 mutation rates ( Table 4 ) . The combination of msh2Δ and dpb3Δ dpb4Δ gave only a two-fold increase in mutation rate on his7-2 reversions and no increase on can1 mutation rates . These data indicate that dpb3Δ dpb4Δ do not act in series with the mismatch repair system . Forward mutations giving resistance to canavanine can arise by many different mechanisms . Earlier studies have shown that even a small collection of sequenced can1 mutants can reveal significant changes in the mutation spectra ( e . g . upon deletion of POL32 , a small subunit of Pol δ , or inactivation of Pol ε proofreading with the pol2-4 allele [9] , [47] ) . To analyze if the deletion of DPB3 and DPB4 might drastically influence the distribution of types of mutations in the CAN1 gene , we sequenced 48 can1 mutants in four isogenic strains: wt , dpb3Δdpb4Δ , pol2-4 and pol2-4 dpb3Δdpb4Δ ( Table 5 ) . The difference between strains carrying the pol2-4 allele and carrying POL2 was statistically significant according to a modified Pearson χ2 test of spectra homogeneity ( see Table S1 and Table S2 ) . There was a characteristic reduction of CG→GC changes and an increase of frameshift mutations in the pol2-4 spectrum ( Table 5 ) . The comparison between wt and dpb3Δ dpb4Δ or pol2-4 and pol2-4 dpb3Δ dpb4Δ showed that the deletion of DPB3 and DPB4 did not give a statistically significant alteration in mutation spectra ( see Table S1 ) . The sample size was insufficient to demonstrate the enhancement of an error signature for Pol ζ despite the increased contribution of REV3 dependent mutations in CAN1 . We conclude that the CAN1 mutations appearing in the pol2-4 background and in the dpb3Δ dpb4Δ mutants arise by different mechanisms . The contribution of Pol ζ to the elevated mutation rates suggested that a Pol2/Dpb2 complex does not support a fully functional replisome . To ask if Dpb3 and Dpb4 influence the processivity of Pol ε , we purified a Pol2/Dpb2 complex with an intact polymerase and exonuclease activity . We measured the processivity of the polymerase activity on a singly-primed , single stranded circular DNA template under single-hit conditions [48] . We found that a 40-fold molar excess of the primer-template over Pol ε and a 20-fold molar excess of the primer-template over Pol2/Dpb2 fit the criteria for single-hit conditions . The processivity of Pol ε on this template was comparable to a previous report , with a strong pause-site 63 nucleotides from the primer ( Figure 1B ) [48] . The absence of Dpb3 and Dpb4 from Pol ε lowers the processivity of polymerization . Some products reached a length of 63 nucleotides , but products were terminated with a higher probability at numerous positions ( Figure 1C ) . On average , the termination probability at each position on the template increased two to three-fold for Pol2/Dpb2 as compared to the four subunit Pol ε . Next , we asked if the Dpb3 and Dpb4 subunits are required for processive exonucleolytic degradation of DNA . We carried out an exonuclease assay with a 57-nt long primer annealed to a 75-nt long template to generate 57-nt dsDNA region . Again , the conditions were empirically determined to achieve single-hit kinetics . This time a five-fold molar excess of primer-template over the four-subunit Pol ε was used , whereas an equimolar ratio of primer-template and enzyme was used for Pol2/Dpb2 . We found that the Pol ε holoenzyme efficiently degraded the first 24 nucleotides of the primer ( Figure 2B ) . At this point only ∼32 nt of the primer remained . This correlates well with the minimal length of dsDNA required for processive synthesis of DNA by Pol ε [32] . By analogy , the processivity of Pol ε exonuclease activity could depend on a specific interaction between the tail-domain and the dsDNA . In agreement with this hypothesis , we found the exonuclease activity of Pol2/Dpb2 to be less processive . Very few primers were degraded further than 11 nucleotides . On average , the termination probability at each position on the primer increased two to three-fold for Pol2/Dpb2 as compared to four-subunit Pol ε ( Figure 2C ) . In addition , the absence of Dpb3 and Dpb4 did not result in a general inactivation of the exonuclease activity , since the exonuclease activity of the Pol2/Dpb2 complex and four-subunit Pol ε was comparable on single-stranded DNA ( data not shown ) . We conclude that Dpb3 and Dpb4 stabilize the interaction of Pol ε with primer-template DNA and therefore positively affect the processivity of the polymerase and exonuclease activities of Pol ε . The removal of Dpb3 and Dpb4 would then lead to frequent dissociation of Pol ε that may disrupt the synthesis of the leading strand and potentially result in single-strand gaps .
The unaltered fidelity of the Pol2/Dpb2 complex suggested that Dpb3 and Dpb4 are not important for the fidelity of DNA synthesis by Pol ε per se . In contrast , our genetic analysis demonstrated that the inactivation of DPB3 and DPB4 in yeast elevates the mutation rates comparable to the proofreading deficient pol2-4 allele of Pol ε . This suggests that the dynamics of the replication fork was altered in the dpb3Δ dpb4Δ strain and the defect influenced the fidelity of the replisome . The hypothesis was supported by the observation that a Pol2/Dpb2 complex ( lacking Dpb3 and Dpb4 ) was less processive both when polymerizing new DNA and degrading DNA ( Figure 1 and Figure 2 ) . Recently , it was shown that Pol ζ participates in the synthesis on undamaged DNA templates during defective replisome-induced mutagenesis as well as synthesis on stretches of single-stranded DNA carrying DNA lesions [42] , [53] . Our genetic analysis supports a role for Pol δ and Pol ζ in spontaneous mutagenesis in dpb3Δ dpb4Δ strains since the mutagenesis in CAN1 depends in part on Pol ζ and Pol δ proofreading suppresses mutations in his7-2 . One explanation for our observations could be that Pol ε dissociates more frequently from the template when DPB3 and DPB4 are deleted . After reinitiation , a gap is left that will be filled in by a post-replication repair mechanism analogous to what might happen when a replicative polymerase encounters a DNA lesion that cannot be bypassed . During this process , there will be time for the 3′-end to repeatedly melt and reanneal . A short homonucleotide run at the his7-2 site may frequently reanneal at the wrong nucleotide creating 1 or 2 nt loops . Such errors could be corrected by proofreading by the replicative polymerases [34] and the presence of Exo+ Pol δ during the gap-filling process would lead to decreased level of mutations . In a pol3-5DV strain , this proofreading is absent leading to a more than additive increase in mutation rate . In this scenario , we detect errors that appear not because of synthetic errors by Pol ε but instead due to an intermediate DNA structure that is prone to frameshift mutations . Thus a greater than additive interaction could be expected because Pol ε without Dpb3 and Dpb4 and proofreading by Pol δ act in series . The effect is detected because of the property of the reversion assay , which focuses on a single mutational pathway . We do not observe the same effects in the CAN1 gene because , in this case , we detect mutations generated by many different pathways . There is , however , a strong synergetic interaction between pol3-5DV and inactivation of mismatch repair . This can easily be explained by the proofreading deficiency of pol3-5DV that generates errors on the lagging strand at the replication fork . In addition , pol3-5DV is a mutator allele due to a defect in Okazaki fragment maturation [54] . Because of the multiple roles of Pol δ and its proofreading activity , more experiments are required to establish the nature of the effect of the pol3-5DV that we observed for his7-2 reversion . The deletion of DPB3 and DPB4 could also result in lesser overall DNA synthesis by Pol ε on the leading strand . This is not likely to be the case as the mutation rate in the pol2-4 strain is not higher than in the pol2-4 dpb3Δ dpb4Δ strain and the mutation signature from pol2-4 in the CAN1 gene is also found in the pol2-4 dpb3Δ dpb4Δ strain , suggesting that exonuclease deficient pol2 synthesize approximately the same amount of DNA regardless if Dpb3 and Dpb4 are present or not . It was earlier shown that defective replicative DNA polymerases ( encoded by pol1-1 , pol2-1 , pol3t and pol3-Y708A ) lead to an increased mutation rate that is in part dependent on Pol ζ . To our knowledge , the in vitro fidelity of the enzymes encoded by the four mutant alleles , pol1-1 , pol2-1 , pol3t and pol3-Y708A has not been measured . Thus , it is not firmly established if these alleles replicate DNA with a reduced fidelity . It is however , plausible that pol3-Y708A has a reduced fidelity based on analogous mutations in the Klenow fragment and RB69 DNA polymerase ( discussed in [44] ) and the position of Tyr708 in the active site [55] . The pol3-t mutant has a temperature sensitive mutation that also may affect the polymerase site and alter the fidelity of Pol3 [44] , [55] . In cases where the effect of mismatch repair has been studied , clear synergy was observed for pol1-1 [56] , pol3-t [57] and pol3-Y708A [44] . Mutant alleles encoding for polymerases that by itself synthesize DNA with a higher error-rate are likely to show synergy with the inactivation of mismatch repair , even if a substantial part of the mutations in the strain are REV3 dependent . The dual mechanism of mutator effects are exemplified by pol3-Y708A , which is likely to encode a polymerase that both generate errors that are corrected by mismatch repair and also induce PCNA ubiquitylation and a Pol ζ dependent increase of mutation rates . Here we present , for the first time , data on a mutant possessing two-subunit Pol ε with a confirmed unchanged error-rate in vitro , and a Pol ζ dependent increase in mutation rates , but no observed synergy with the inactivation of mismatch repair . This observation provides a distinction of errors made at the replication fork from errors made during postreplication DNA synthesis . The deletions of DPB3 and DPB4 led to an increased mutation rate but did not act in series with msh6Δ , msh2Δ , pms1Δ or mlh1Δ . The lack of synergy could be due to an essential function for DPB3 or DPB4 in mismatch repair that inactivates the mismatch repair system . It was proposed earlier that the 3′→ 5′exonuclease activity of Pol ε could be involved in the excision step of mismatch repair in yeast [35] . However , the reversion rate at the lys2::insE-A14 allele in the dpb3Δ dpb4Δ strain is too low to support a role for DPB3 and DPB4 in mismatch repair ( compare dpb3Δ dpb4Δ with msh6Δ , msh2Δ , pms1Δ or mlh1Δ ( Table 1 and Table 4 ) ) . Yet , there is a possibility that redundancy , due to genes with over-lapping functions in mismatch repair , suppress the mutation rate in dpb3Δ dpb4Δ strains . The possible redundancy only allows us to conclude that there is no evidence for a role of DPB3 and DPB4 ( or Pol ε ) in mismatch repair . Whether mismatch repair is carried out in the near proximity of the replication fork or is uncoupled from the replication fork remains unclear . It has been proposed that mismatch repair may be physically linked to the replication fork [58] , but DNA lesions from MNNG may induce a futile repair cycle where mismatch repair functions outside the S-phase [59] . Based on the genetic analysis of DPB3 and DPB4 we propose a model with two zones where mutagenesis occurs during DNA replication . The first zone is in the near proximity of the replication fork where Pol ζ- independent mutagenesis occurs and errors are corrected by mismatch repair . The second zone where Pol Δ and Pol ζ carries out post-replication repair is uncoupled from the replication fork . In this zone , Pol ζ-dependent mutagenesis occurs and errors are not at all or very inefficiently corrected by the mismatch repair system . There is a series of observations upon which this model is based . We have clearly shown that the mutagenesis of the CAN1 gene in dpb3Δ dpb4Δ strains depends on Pol ζ and some mutation intermediates in the his7-2 allele are proofread by Pol Δ . Gap-filling during a post-replication repair process is likely to depend on PCNA , thus giving an advantage to Pol δ and Pol ζ over Pol ε to synthesize DNA since Pol ε has a slow on-rate on the PCNA-primer ternary structure [14] , [48] . Recently , during defective-replisome-induced mutagenesis it was independently shown that Pol ζ replicates undamaged DNA under conditions when the dynamics of the replication fork is affected [42] . Mismatch repair is functional in the dpb3Δ dpb4Δ background and yet there is no synergism . A hypothetical role for Rev3 in the close proximity of the replication fork would result in replication errors that are expected to be corrected by the mismatch repair system , analogous to errors produced by proofreading deficient Pol ε . At this position Rev3 would be a major contributor to the mutation rate in a msh2Δ strain , because Rev3 is responsible for at least half the mutation rate in the CAN1 gene in wild-type strains . The combination of rev3Δ and msh2Δ would then result in a substantially lower mutation rate; however , this was not the case . Instead , the mutation rate in a rev3Δmsh2Δ strain was comparable to a msh2Δ strain [47] , suggesting that errors by Pol ζ are not efficiently corrected by mismatch repair . Our results demonstrate that errors generated by Pol ζ are not efficiently repaired by mismatch repair and are supported by evidence that some mutations generated by Pol ζ in a rad52Δ background are not corrected by mismatch repair in the lys2Δ A746-NR allele [60] . Based on the sum of these observations we propose that the deletion of DPB3 and DPB4 results in a decreased Pol ε processivity , generating a DNA substrate , which must be processed to some extent by Pol δ and Pol ζ during post-replication repair . This event occurs in a zone separated from active replication forks where the correction of errors by mismatch repair may be inefficient . Alternative interpretations could be that synthesis by Pol ζ at stalled replication forks is not under mismatch repair surveillance . The transient dissociation of Pol ε would , in this case , create specific conditions when mismatch repair cannot function . Under these specific conditions any repair synthesis at the replication fork as well as post replicative repair synthesis in the single stranded gaps might escape MMR . The influence of chromatin on mutation rates and mismatch repair could also be an explanation . A potential mechanism could be that PCNA is post-translationally modified , due to check-point activation , blocking the interaction between mismatch repair proteins and PCNA . Regions with post-translationally modified PCNA would then be less efficiently repaired by mismatch repair . Mismatch repair genes are not exclusively involved in correcting replication errors at the replication fork . Recently , it was shown that mismatch repair genes suppress recombination and promote translesion synthesis by Pol ζ in an assay measuring spontaneous mutation rates [60] . Other examples are immunoglobulin genes where mismatch repair together with Pol η is required for hypermutation at A/T pairs [61] . This is a paradox as the mismatch repair system promotes error-prone DNA synthesis by Pol ζ and Pol η in these two examples . The deletion of DPB3 and DPB4 unveils another example of how error-prone DNA synthesis is accepted to complete DNA synthesis and the mismatch-repair system does not correct the errors . The contribution of these Pol ζ dependent errors is small when compared to the error load which is corrected by mismatch repair at the proximity of the replication fork ( compare CAN1 mutation rate in dpb3Δ dpb4Δ with msh2Δ ( Table 4 ) ) . Yet , we found that the error-rate in the dpb3Δ dpb4Δ strain was comparable to the pol2-4 strain . The error-rate in pol2-4/pol2-4 mice was recently reported to be sufficient to support tumor development in mice [62] . Although the mechanism by which the error rates increases in pol2-4 and dpb3Δ dpb4Δ strains clearly differs , it is tempting to speculate that the inactivation of the mammalian homologues to DPB3 and DPB4 could result in defective replisomes , elevated mutation rates and tumor development .
All S . cerevisiae strains used in this study are isogenic to E134 ( MATα ade5-1 lys2::InsEA14 trp1-289 his7-2 leu2-3 , 112 ura3-52 ) [33] . The dpb3Δ mutant was kindly provided by P . Shcherbakova and is described in [27] . Other strains carrying dpb3Δ were obtained as described in [27] . The dpb4Δ mutants were constructed by transformation with PCR fragment carrying the hygB selectable marker and obtained using primers DPB4/kanMX-F ( 5′-ATGCCACCAAAAGGTTGGAGAAAAGACGCCCAAGGGAATTACCCCCGTACGCTGCAGGTCGAC ) and DPB4/kanMX-R ( 5′-TTACGTTTGCTCAAGGTTTTGAACTCTAGTTTCTACATCTTGGCTATCGATGAATTCGAGCTCG ) and the pAG32 plasmid as a template [63] . The disruption was confirmed by PCR analysis . The pol2-4 mutation was obtained as described in [64] using YIpJB1 plasmid carrying pol2-4 mutation [65] . The pol3-5DV mutation was obtained as described in [64] using plasmid p170-5DV [66] . The presence of the pol2-4 mutation after integration into the chromosome was confirmed by SfcI digest of short PCR fragment encompassing mutation and DNA sequencing . Deletion of the MSH6 gene was obtained as described in [67] . The REV3 gene , encoding the catalytic subunit of Pol ζ , was deleted as described in [68] . Deletion of the MSH2 gene was obtained by transformation with PCR product obtained from the pRS305 plasmid using oligos MSH2_del_F ( CTCCACTAGGCCAGAGCTAAAATTCTCTGATGTATCAGAGGAGAGCAGAGCAGATTGTACTGAGAGTGCACC ) and MSH2_del_R ( CCTTCACTTTTCTAATCCACTCTTTCAGTAAAGCCTTCAAACGAACGCATCTGTGCGGTATTTCACACCGC ) . The same strategy was used for deletion of the PMS1 and MLH1 gene . To obtain pms1Δ we used oligos PMS1_A ( TATCAAAGCTAGATCATATTTCGTAATCCTTCGAAAATGAGCTCCAATCACGTAAAATATCTTGACCGCAGTTAA ) and PMS1_S ( AAGGTGTAAGCAAAAGGAACAGAGGTATATCCCTGTGAAATATTTATTTAGCCCCTATGAACATATTCCATT ) . The mlh1Δ was obtained by transformation with the PCR product obtained from the pRS306 plasmid using oligos MLH1_A ( AAGTTAACACCTCTCAAAAACTTTGTATAGATCTGGAAGGTTGGCTATTTCCAACACCGCAGGGTAATAACTGAT ) and MLH1_S ( ATACGATAGTGATAGTAAATGGAAGGTAAAAATAACATAGACCTATCAATAAGCACGGTCACAGCTTGTCTGTAA ) . The fluctuation tests to determine spontaneous mutation rates were , unless otherwise indicated , performed in two to five independent experiments of nine independent cultures each with independently obtained derivatives . Single two-day-old colonies from YPD plates were inoculated in 5 ml of liquid YPD medium and were grown with strong aeration for two days and processed as described [33] . Independent His+ revertants and Canr mutants were grown as small patches on YPD plates . Regions of corresponding genes were amplified by PCR . Amplified DNAs were purified by QIAGEN PCR purification kit and sequenced by MWG Biotech ( www . mwgdna . com ) . CAN1 spectra obtained in four strains were compared using several statistical techniques . A Monte Carlo modification of the Pearson χ2 test of spectra homogeneity [69] was used to compare 2 x N tables ( two mutation spectra , N≥2 ) . Small probability values ( P≤0 . 05 ) indicate a significant difference between two spectra . Calculations were done using the program COLLAPSE [70] . All purification steps were carried out as described in [32] . We used primer 3NY ( 5′-AGGTCACGATGCGGCATAGCCTGCATTGATCGCACGATGATCAGCGGACTGCTTACC ) annealed to the template 19wt ( 3′-TCCAGTGCTACGCCGTATCGGACGTAACTAGCGTGCTACTAGTCGCCTGACGAATGGACAGTGCCATTGTCACTG ) as a substrate for the exonuclease reaction . The primer ( 8 µM ) was labeled with 40 µCi of [γ-32P]ATP in a 20-µl reaction with 10 U T4 polynucleotide kinase ( Promega ) for 1 h . The reaction was stopped with EDTA and labeled products were purified through PAGE . The end-labeled primer was annealed to the template at 1 . 5: 1 ratio for a 5 minute incubation at 80°C followed by slow cooling to room temperature . For the exonuclease assay 0 . 5 nM substrate was incubated with 0 . 1 nM Pol ε or 0 . 5 nM Pol2/Dpb2 complex in a 65 µl reaction mixture ( 40 mM Tris-HCl pH 7 . 8; 1 mM DTT; 0 . 2 mg/ml Ac-BSA; 8 mM MgCl2; 125 mM NaAc ) . Fifteen µl aliquots were taken at the indicated time points and were mixed with 8 µl stop solution ( 80% formamide; 50 mM EDTA; 1 mM bromophenol blue ) . Before loading the reactions on a 12% polyacrylamide-urea gel , the primer-templates were denatured at 99° C for 4 min and then cooled on ice . The intensity of bands corresponding to different exonuclease products was quantified using phosphoimager plates and the ImageQuant software package supplied with a Typhoon 9400 phosphoimager ( Amersham Biosciences ) . For primer extension assay a [γ-32P]ATP –labeled ( as described previously ) 50-mer oligonucleotide was annealed to the pBluescript II SK ( + ) ssDNA in a ratio of 1∶1 . 5 . For the DNA synthesis processivity assay , the substrate ( 14 nM ) was incubated with the four-subunit Pol ε ( 0 . 35 nM ) or Pol2/Dpb2 complex ( 0 . 7 nM ) in a reaction mixture ( 40 mM Tris-HCl pH 7 . 8; 1 mM DTT; 0 . 2 mg/ml Ac-BSA; 8 mM MgCl2; 125 mM NaAc; 100 µM dNTP ) . Because the loading efficiency of the Pol2/Dpb2 complex on DNA was compromised we used a two and five-fold higher concentration of the Pol2/Dpb2 complex , compared to the full-subunit Pol ε , in the primer-extension assay and exonuclease assay , respectively . The conditions were empirically determined to meet single-hit kinetics , i . e . where a polymerase molecule never re-associated with a previously extended primer . In the primer-extension assay , the termination probability at position N at each primer/template was calculated by dividing the intensity of the band N by the intensity of all bands ≥ N . In the exonuclease processivity assay , the termination probability at position N at each primer/template was calculated by dividing the intensity of the band N by the intensity of all bands ≤ N [71] . DNA synthesis fidelity was measured using the bacteriophage M13mp2 forward mutation assay described previously [39] , [40] . Briefly , double-stranded M13mp2 DNA with a 407-nucleotide single-stranded region containing a portion of the lacZ gene was used as a substrate for in vitro DNA synthesis . Reactions mixtures contained ∼1 . 5 nM DNA template , 50 mM Tris-Cl ( pH 7 . 5 ) , 2 mM DTT , 100 µg/ml BSA , 10% glycerol , 250 µM dNTPs and 14 nM wild type or exonuclease-deficient 2-subunit Pol ε . Reactions were incubated at 30°C for 30 min . Aliquots of the reactions were analyzed by agarose gel electrophoresis to confirm complete gap-filling , and another aliquot of DNA was introduced into E . coli to score the frequency of light blue and colorless plaques reflecting errors made during in vitro DNA synthesis . Single stranded DNA was isolated from independent mutant M13 plaques and the lacZ gene was sequenced . Error rates ( ER ) for individual types of mutation were calculated according to the following equation: ER = [ ( Ni/N ) ×MF]/ ( D×0 . 6 ) where Ni is the number of mutations of a particular type , N is the total number of mutants analyzed , MF is frequency of lacZ mutants , D is the number of detectable sites for the particular type of mutation , and 0 . 6 is the probability of expressing a mutant lacZ allele in E . coli . | The high fidelity of DNA replication is safeguarded by the accuracy of nucleotide selection by DNA polymerases , proofreading activity of the replicative polymerases , and the DNA mismatch repair system . Errors made by replicative polymerases are corrected by mismatch repair , and inactivation of the mismatch repair system results in a multiplicative increase in error rates when combined with a proofreading deficient allele of a replicative polymerase . In this study , we demonstrate that the deletion of two non-essential genes encoding for two subunits of Pol ε give an increased mutation rate due to increased synthesis by the error-prone DNA polymerase ζ . Surprisingly , there was no multiplicative increase in error rates when the mismatch repair system was inactivated . We propose that the deletion of DPB3 and DPB4 gives a defective replisome , which in turn gives increased synthesis , in part , by Pol ζ during an error-prone post-replication process that is not efficiently repaired by the mismatch repair system . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"molecular",
"biology/dna",
"repair",
"biochemistry/replication",
"and",
"repair",
"molecular",
"biology/dna",
"replication",
"genetics",
"and",
"genomics/chromosome",
"biology"
] | 2010 | Mismatch Repair–Independent Increase in Spontaneous Mutagenesis in Yeast Lacking Non-Essential Subunits of DNA Polymerase ε |
In a murine model of moderate childhood malnutrition we found that polynutrient deficiency led to a 4–5-fold increase in early visceralization of L . donovani ( 3 days post-infection ) following cutaneous infection and a 16-fold decrease in lymph node barrier function ( p<0 . 04 for all ) . To begin to understand the mechanistic basis for this malnutrition-related parasite dissemination we analyzed the cellularity , architecture , and function of the skin-draining lymph node . There was no difference in the localization of multiple cell populations in the lymph node of polynutrient deficient ( PND ) mice , but there was reduced cellularity with fewer CD11c+dendritic cells ( DCs ) , fibroblastic reticular cells ( FRCs ) , MOMA-2+ macrophages , and CD169+ subcapsular sinus macrophage ( p<0 . 05 for all ) compared to the well-nourished ( WN ) mice . The parasites were equally co-localized with DCs associated with the lymph node conduit network in the WN and PND mice , and were found in the high endothelial venule into which the conduits drain . When a fluorescent low molecular weight ( 10 kD ) dextran was delivered in the skin , there was greater efflux of the marker from the lymph node conduit system to the spleens of PND mice ( p<0 . 04 ) , indicating that flow through the conduit system was altered . There was no evidence of disruption of the conduit or subcapsular sinus architecture , indicating that the movement of parasites into the subcortical conduit region was due to an active process and not from passive movement through a leaking barrier . These results indicate that the impaired capacity of the lymph node to act as a barrier to dissemination of L . donovani infection is associated with a reduced number of lymph node phagocytes , which most likely leads to reduced capture of parasites as they transit through the sinuses and conduit system .
Protein-energy malnutrition ( PEM ) is thought to be the most frequent cause of human immunodeficiency [1] , and greatly predisposes individuals to infectious diseases in resource-poor regions of the world [2] . In its synergy with infection , under-nutrition contributes to approximately 50% of childhood deaths worldwide [3] . Apart from PEM , deficiencies in single nutrients , such as vitamins , fatty acids , amino acids and trace elements also alter immune function and increase the risk of infection [2] . Both innate and adaptive immunity may be impaired in the malnourished host [4] , leading to increased susceptibility to infectious diseases [5] . Malnutrition also impairs the development of a normal immune system during the critical periods of pregnancy , neonatal maturation , and weaning [6] , [7] . Inadequate intake of dietary energy and protein leads to atrophy and alteration in the architecture of lymphoid organs , such as the thymus and spleen . Severe thymic atrophy results from massive thymocyte apoptosis ( particularly affecting the immature CD4+CD8+ cell subset ) and decreased cell proliferation . In the spleen , there is loss of lymphoid cells around the small blood vessels [8]–[10] . However , the influence of malnutrition on the lymph node architecture and function has not been studied . One of the infections whose risk is increased by malnutrition is visceral leishmaniasis ( VL ) , caused by the intracellular protozoan parasites of the Leishmania donovani complex ( L . infantum ( L . chagasi ) and L . donovani ) . VL is a significant health problem in resource-poor regions of the world , particularly in India , Sudan , Bangladesh , and Brazil [11] , [12] . Following inoculation of the parasite in the dermis by the bite of an infected sand fly the parasite disseminates to infect phagocytic cells of spleen , liver and bone marrow . The majority of individuals who are infected with L . donovani or L . infantum are able to control the infection and develop a sub-clinical asymptomatic infection; a minority ( usually <10% ) of infected individuals develops severe hepatosplenomegaly , fever , pancytopenia , and cachexia which ultimately progresses to death unless the patient is treated [13]–[16] . The factors that influence susceptibility to leishmaniasis and its progression are incompletely understood , but several lines of evidences suggest that malnutrition is a primary risk factor that contributes to the development of VL in children . Epidemiologic studies have documented an increased risk for VL in the malnourished host [2] , [17]–[20] and children with moderate to severe PEM were found to have about a nine-fold increased risk of developing VL [18] . Malnutrition was identified as a risk factor for severe disease [18] and death from VL in both children ( WFH<60%; OR 5 . 0 ) and adults ( BMI<13; OR 11 . 0 ) [21] . Malnutrition-related VL is particularly evident in displaced and impoverished populations [22] . The mechanistic relationship between malnutrition and the course of VL at the molecular and cellular level is poorly understood . A better understanding of those mechanisms might offer new opportunities for prevention or therapeutic dietary intervention . Moreover , understanding the interplay of nutrients and immune function is of additional interest because of the malnutrition- related risk of infection with other pathogens . In previous experimental studies in a murine model of malnutrition that mimicked the growth characteristics of human weanling malnutrition [23] , we observed that malnutrition caused a failure of lymph node ( LN ) barrier function that led to a profound increase in the early ( 3 days post-infection ) dissemination of L . donovani to the visceral organs ( liver and spleen ) . The function of the lymph node as a barrier to delay or reduce pathogen dissemination is a function of the capture of the pathogen by phagocytes ( largely macrophages and dendritic cells ) and restriction of its transit through the node via the size-exclusion properties of the node architecture [24] , [25] . This barrier function then enables development of innate and adaptive immune responses that effect killing of the pathogen . The concept of LN barrier function has been widely discussed in the prevention of tumor metastases [26] , but is less studied in the field of infectious diseases . In the work presented here , using the murine model that we established previously [23] , we further investigated the mechanisms by which polynutrient ( protein , iron and zinc ) deficiency ( PND ) impaired the capacity of the LN to act as a barrier to dissemination of L . donovani infection . We found that PND reduced the mass and cellularity of the LN , particularly affecting fibroblastic reticular cells and myeloid phagocytic cells , without disrupting the overall LN architecture . The function of the LN reticular conduit system was also altered and parasites were found to be associated with the conduit in the LN subcortical region and within the high endothelial venule , into which the conduits drain . The reduced number of LN phagocytes , which would affect the overall phagocytic capacity of the organ , and the altered function of the LN conduit system , therefore are likely contributors to the reduced retention and increased escape of parasites or parasitized host cells from the LN to the visceral organs .
This study was carried out in strict accordance with the recommendations in the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health . The protocol was approved by the Institutional Animal Care and Use Committee of the South Texas Veterans Health Care System where all animal experimentation was conducted . Weanling ( 3-week-old ) female BALB/c mice were obtained from Charles River Laboratories , Inc . ( Wilmington , MA ) . Mice were maintained in specific pathogen-free conditions at the Veterinary Medical Unit of the Department of Veterans Affairs Medical Center , South Texas Veterans Health Care System ( STVHCS ) , San Antonio , TX . Mice were initially weight-matched and housed as four mice per cage in standard polycarbonate shoebox cages with low trace element bedding ( Alpha-Dri; Shepard Specialty Papers , Kalamazoo , Mich . ) . The mice had free access to water and were acclimatized to standard laboratory mouse chow ( Teklad LM-485; Harlan Teklad , Madison , WI ) for three days prior to initiation of the two experimental diets . The well-nourished ( WN ) control group of mice received a diet of normal mouse chow with 17% protein , 100 ppm iron , 30 ppm zinc ( Teklad ) , which was provided ad libitum . The PND mice received a diet of mouse chow identical to the normal control diet except for low protein ( 3% ) , iron ( 10 ppm ) , and zinc ( 1 ppm ) ( Teklad ) , as previously described [23] . The PND mice received 90% of the weight of food consumed per day by the mice in the WN group to ensure that they did not increase their consumption in response to the nutrient deficiencies , which resulted in approximately a 10% reduction in caloric intake . Mice were fed the experimental diets until the completion of the experiment ( 28–31 days ) . The body weights of the mice were measured once per week , and food intake was recorded on a twice-weekly basis in order to calculate the amount of chow to provide to the PND group on subsequent days . Blood was collected from the mice by terminal cardiac puncture . After clotting and centrifugation the serum was collected and stored at −80°C until use . Liver tissue was collected following exsanguination of mice and stored at −80°C until use . Serum albumin levels and liver iron and zinc levels were determined by automated photometry at the Texas Veterinary Medical Diagnostic Laboratory , College Station , Texas . Leishmania donovani ( 1S strain; MHOM/SD/00/S-2D ) promastigotes were grown in complete M199 medium for 6 days and the metacyclic forms were isolated as described previously [27] . The virulence of the parasites was maintained by regular isolation from spleen tissue from infected mice or hamsters . Mice that had received the experimental diet for 28 days were inoculated with 106 metacyclic promastigotes in 20 µl Dulbecco's Modified Eagle Medium ( DMEM ) in the skin over each hind foot-pad . In some experiments mice were infected with 2×106 parasites that had been labeled with the membrane fluorescent dye PKH26 ( Sigma-Aldrich , St . Louis , MO ) as described previously [28] . At 3-days post-infection , the infected mice were euthanized and the popliteal lymph nodes , spleen , and liver were harvested and weighed . Real-time quantitative PCR ( qPCR ) targeting kinetoplast DNA was used to quantify L . donovani amastigotes in the homogenized tissues as described previously [29] . Briefly , LN , spleen and liver tissues were homogenized in phosphate-buffered saline ( PBS ) at 1 mg/10 µl and 100 µl of the homogenate was used for DNA extraction ( Qiagen DNA extraction kit ) . Forty ng of the extracted DNA was amplified with the Abi Prism 7900 using real-time PCR master mix kit ( Applied Biosystems ) , 400 nM of the 13A and 13B primers [29] , and 100 nM of the internal probe ( 5′- ( 6-FAM ) -TTGAACGGGATTTCTGCACCCA- ( TAMRA ) -3′ ) . To quantify the number of parasites , a standard curve was generated by amplification of 10-fold dilutions of L . donovani amastigotes isolated from the same tissue in a separate reaction . The parasite concentration was calculated per milligram of tissue , and the total organ parasite burden was calculated by multiplying this concentration by the whole-organ weight [23] . Lymph nodes were collected in RPMI media supplemented with 2% fetal bovine serum ( FBS; Gibco ) . Lymph node cell suspensions were prepared by cutting the tissue into small pieces and digesting it for 30 min at 37°C with collagenase D ( Roche ) at 2 mg/ml in buffer containing ( 150 mM NaCl , 5 mM KCl , 1 mM MgCl2 , 1 . 8 mM CaCl2 , 10 mM Hepes pH 7 . 4 ) . The tissue was further minced and strained through 100-µm cell strainers ( Becton Dickinson [BD] , San Jose , CA ) , and washed once in a solution of PBS with 2% FBS and 0 . 1% sodium azide . The cells were resuspended in 500 µL of RPMI with 2% FBS . The cells were counted and adjusted at a concentration from 100 , 000 to 500 , 000 cells per 50 µl , incubated for 15 min with 0 . 8 µg FC block at room temperature , followed by the relevant antibodies for 30 minutes at room temperature in the dark , washed again in PBS with FBS and azide and finally fixed in 250 µl of FACS lysing solution ( BD , Biosciences ) . Cell surface analysis was performed using a combination of a panel of surface markers: FITC-conjugated rat anti-mouse Ly-6G , clone 1A8 , PE-conjugated hamster anti-mouse CD11c , clone HL3 , PE-conjugated rat anti-mouse Ly-6G and Ly-6C , FITC-conjugated rat anti-mouse CD11b , mouse T lymphocyte subset antibody cocktail ( PE-conjugated rat anti-mouse CD4 , PE-Cy7 conjugated rat anti-mouse CD3e , and FITC conjugated rat anti-mouse CD8 ) , FITC conjugated rat anti-mouse CD45R/B220 , clone RA3-6B2 ( BD PharMingen , San Diego , CA ) , rat anti-mouse CD169 , clone 3D6 . 112 ( Abcam ) , FITC conjugated rat antimouse CD169 , clone 3 D6 . 112 , FITC conjugated rat anti-mouse CD31 , and Alexa-Fluor 488 and 647 conjugated hamster anti-mouse CD11c , clone N418 ( AbD Serotec , Raleigh , NC ) . For intracellular MOMA-2 analysis , cell preparations were fixed and permeabilized with fixation/permeabilization buffers ( AbD Serotec ) and stained with Alexa Fluor 488-conjugated rat antimouse macrophage/monocytes ( AbD Serotec ) . For intracellular ER-TR7 analysis , cell preparations were fixed and permeabilized with a mixture of ethanol/acetone ( 7∶3 ) and stained with PE , FITC-conjugated rat antimouse ER-TR7 ( Santa Cruz Biotechnology , Santa Cruz , CA ) , or unlabeled rat antimouse ER-TR7 ( Abcam ) . Appropriate rat or hamster IgG isotype antibodies were used as controls . The secondary antibody APC-conjugated goat anti-rat IgG ( Santa Cruz Biotechnology , inc ) was used in the indirect staining . All flow cytometric analyses were performed on a FACSAria flow cytometer ( Becton Dickinson , San Jose , CA , USA ) . The dissected popliteal lymph nodes were fixed in 10% neutral buffered formalin and processed routinely into paraffin . The fixed paraffin-embedded tissues were sectioned at 3–4 µm and stained with hematoxylin and eosin ( H&E ) . Some paraffin embedded sections were used for reticulin stain using the method of Gomori [30] . Popliteal lymph nodes were fixed in 4% paraformaldehyde plus 1% glutaraldehyde and processed for plastic embedment using conventional methods . Thin sections ( 60 to 70 nm ) were stained with lead citrate and uranyl acetate . The lymph node conduit network was examined and photographed using a Jeol-JEM-1230 transmission electron microscope ( Tokyo , Japan ) . Popliteal lymph nodes were removed from mice , immediately embedded in Tissue Tek Optimum Cutting Temperature compound ( Sakura FineTek , Torrance , CA ) , and snap frozen and stored at −80°C until used . Sections ( 6 µm in thickness ) were cut in a cryostat and placed on positively charged microscope slides ( Superfrost/Plus , Fisher Scientific ) . Sections were air-dried overnight and fixed for 10 minute in ice-cold acetone . Sections were blocked with 10% serum and stained with antibodies diluted in 2% serum , which was from the same species in which the secondary antibodies were raised . Primary and secondary antibodies were applied for 60 min at room temperature in a humidified chamber . Slides were washed between and after antibody applications 5 times with PBS/0 . 02% BSA for 5 min each . Slides were coverslipped with Gold Prolong anti-fade mounting media ( Molecular Probes , Eugene , OR ) . The antibodies used , and their source and specifications are summarized in Table 1 . Stained lymph node sections were examined using an Olympus Provis AX 70 fluorescent microscope . The image-proplus software ( Media Cybernetics , Inc . , Bethesda , MD ) was used to count the intensity of the fluorescence as a proportion of tissue area . For evaluation of conduit function , the skin over each footpad was injected with 20 µl low molecular weight ( 10 kD ) or high molecular weight ( 500 or 2000 kD ) lysine fixable Texas Red- or FITC-labeled dextran ( 6 mg/ml , Invitrogen , Grand Island , NY ) . Mice were euthanized 3 minutes after injection by CO2 asphyxiation and the draining popliteal lymph nodes were harvested . The lymph nodes were immediately placed in freshly prepared 4% paraformaldehyde ( pH 7 . 4; room temperature for 1–2 h , then 4°C for 2 h ) , washed twice in PBS and saturated overnight at 4°C in 30% sucrose before being embedded in Tissue-Tek optimum cutting temperature compound . The sections ( 6 µm ) were fixed in acetone for one minute and stained and visualized by fluorescence microscopy as described above . Data are expressed as the mean ± SEM and were analyzed using Prism software ( GraphPad , La Jolla , CA ) . The parametric unpaired t test , or the non-parametric Mann-Whitney U test were used for the analysis depending on the normalcy of distribution of the data . Data were considered statistically significant if p≤0 . 05 .
WN mice received a normal diet ( 17% protein , 100 ppm Fe , and 30 ppm Zn ) ad libitum and consumed approximately 2 . 1 g of food per day . PND mice received a diet deficient in protein ( 3% ) , Fe ( 10 ppm ) , and Zn ( 1 ppm ) and received 90% of the quantity of food consumed by the WN mice ( approximately 1 . 9 g per day ) . After 4 weeks of feeding the experimental diets , the PND mice showed a slightly slumping growth curve with a 15 . 3% average reduction from baseline weight after 28 days ( Fig . 1A ) , which was consistent with the previous report that showed it was comparable to moderate human weanling malnutrition [23] . To evaluate the nutritional status of the mice , the concentrations of serum albumin and hepatic zinc and iron were determined in PND mice 28 days after initiating the experimental diet . We observed a significant reduction in the serum albumin concentration ( Fig . 1B; p = 0 . 006 ) and zinc and iron concentration in the liver of PND compared to WN mice ( Fig . 1C , 1D; p = 0 . 02 and p = 0 . 01 , respectively ) . To investigate the effect of malnutrition on the early visceralization of L . donovani , PND and WN mice were inoculated in the skin over the hind footpad with 106 L . donovani metacyclic promastigotes , and the parasite burdens in liver , spleen and draining ( popliteal ) lymph node were determined at 3 days post-infection . Consistent with our previous observations , in four different experiments using a lower parasite inoculum and shorter period of dietary deficiency than what we had described previously [23] , the parasite burden in the lymph node ( calculated as either the number of parasites per mg tissue or as the total organ parasite burden ) was lower in the PND mice than in the WN group ( Fig . 2A , 2B ) . Notably , the PND mice showed greater L . donovani dissemination to spleen ( Fig . 2C , 2D ) and liver ( Fig . 2E , 2F ) compared with the WN group . In 3 independent experiments the total measured extradermal parasite burdens ( parasite burdens in LN+spleen+liver ) showed no difference between the two groups of mice ( for the experiment shown in Fig . 2 the total number of parasites for the WN and PND mice was 98 , 403±31 , 253 and 93 , 656±16 , 127 , respectively , p = 0 . 9 ) , which indicated that there was no difference in the parasite survival between the WN and the PND mice at this early stage of infection . However , the total visceral parasite burden was higher in the PND group than the WN group and the total lymph node parasite burden was higher in the WN group than the PND group , which together led to a 16-fold reduction in the calculated percent lymph node barrier function [23] in the PND infected mice ( Fig . 2G ) . It is generally accepted that Leishmania traffic from the skin to the draining LN through the afferent lymphatics [31] . However , malnutrition could alter that route of transit by facilitating increased entry of the parasites directly into the bloodstream from the skin , thereby bypassing the LN . To address this possibility we quantified parasites in the skin , draining LN and visceral organs at a much earlier time point ( 16 hrs post-infection ) . We found no difference in the number of parasites in the skin or LN ( Fig . S1 ) , suggesting that malnutrition did not lead to increased visceralization by the parasite bypassing of the draining LN early in the infection process . Otherwise the parasite burden would have been reduced in both the skin and LN in the PND compared to WN mice . Collectively , these data support our previous work [23] that suggested that malnutrition produced increased visceralization after cutaneous L . donovani infection due to the failure of the draining lymph node to act as a barrier to dissemination . It was reported that lymphoid organs such as the thymus and spleen showed significant atrophy in patients with PEM or zinc deficiency [32] , [33] , but the effect on the LN had not been investigated . To determine the effect of polynutrient deficiency on LN mass and cellular composition , the popliteal LNs from groups of WN and PND mice were harvested before or 3 days after L . donovani infection . The lymph node weights were significantly less in the PND groups , whether they were infected or uninfected , compared to their WN counterparts ( Fig . 3A ) . When corrected for body weight ( LN weight index = LN weight/body weight ) , there was no difference in the relative weights of the LNs from the uninfected WN and PND mice , however , the LN weight index was significantly lower in the infected PND compared to infected WN mice ( Fig . 3B ) . Consistent with the reduced LN mass we found a decrease in total LN cell number in PND compared to WN mice , regardless of whether or not they had been challenged with L . donovani ( Fig . 3C ) . In both the WN and PND mice there was approximately a 6-fold increase in LN cell number at 3 days after L . donovani infection ( Fig . 3C ) . Histological examination of the LN of the PND mice similarly revealed a marked decrease in the size of the LN , however , there was no obvious difference in the gross histopathology observed in hematoxylin and eosin ( H&E ) stained LNs of PND group compared with the WN groups ( data not shown ) . Thus , although malnutrition caused a generalized reduction in LN mass and cellularity , it did not appear to alter the gross structure of the LN . To further evaluate the effect of malnutrition on the cellular composition of the lymph node , we examined cell populations in the draining lymph node of L . donovani-infected and uninfected WN and PND mice . We focused on the cell populations that might be involved in transporting the parasite from the site of cutaneous infection to the draining lymph node ( generally ascribed to dendritic cells ) , as well as neutrophils , macrophages , and fibroblastic reticular cells ( FRC ) , which may play a role in internalization of Leishmania in the lymph node [31] , [34] . In uninfected mice , by flow cytometry we found that there was no difference in the percentage of dendritic cells ( CD11c+ ) in the PND and WN groups ( Fig . 4A , left panel ) , but the PND mice had a reduced total number of CD11c+ cells compared to WN controls ( p = 0 . 0002 ) ( Fig . 4A , right panel ) . This was confirmed by immunofluorescence staining of tissue sections from uninfected mice ( Fig . 4G ) . Although we did not detect any difference in the percentage of macrophages ( MOMA-2+ or CD11c−CD11b+ ) , FRCs ( ER-TR7+ ) , subcapsular sinus ( SCS ) macrophages ( CD169+ cells that line the floor of the LN subcapsular space and medulla [35] ) , or neutrophils ( GR1+Ly6G+ ) between uninfected PND and WN mice ( Figs . 4B–4F , left panels ) , the total number of CD11c−CD11b+ cells , ER-TR7+ cells , and CD169+ cells were significantly reduced in the uninfected PND compared to the uninfected WN group ( Figs . 4C , 4D , 4E , right panels; p = 0 . 008 , p = 0 . 008 and p = 0 . 004 , respectively ) . A similar reduction in macrophages and DCs was also evident in the spleens of uninfected PND compared to WN mice ( Fig . S2; p<0 . 001 ) indicating that malnutrition also had an effect on myeloid cells in organs other than the LN . No statistical difference in the total number of MOMA-2+ macrophages or neutrophils ( Gr1+Ly6+ ) was found between the uninfected PND and WN groups ( Figs . 4B , 4F , right panels ) and immunofluorescence revealed no difference in endothelial cells ( CD31+ ) , B cells ( B220+ ) and T cells ( CD3+ ) ( Fig . 4G ) . When draining LNs from mice challenged with L . donovani were examined , greater quantitative differences in the LN cell populations became evident . Independent of nutritional status , L . donovani infection dramatically expanded the populations of all cell types in the LN ( 4–47 fold-increase for WN mice and 2–75 fold-increase for PND mice; Figs . 4A–4F ) . In the popliteal LNs of infected PND mice compared to infected WN mice we found a reduced total number ( p<0 . 0001 ) but not percentage of CD11c+ cells ( Fig . 4A , right panel ) , but macrophages ( MOMA-2+ or CD11c−CD11b+ ) were reduced in both percentage ( p = 0 . 03 and p = 0 . 01 , respectively ) and total number ( p = 0 . 0002 and p = 0 . 0003 , respectively ) in the PND infected mice ( Figs . 4B and 4C ) . There was no difference in the percentage of FRC ( ER-TR7+ ) in the infected WN and PND groups , but a significantly reduced total number of FRC was found in the PND mice ( Fig . 4D; p = 0 . 01 ) . CD169+ SCS macrophages also showed a reduced percentage and total number ( Fig . 4E; p = 0 . 04 and p = 0 . 003 , respectively ) in the PND infected mice , and this was corroborated by immunofluorescence ( Fig . 4H ) . The total number , but not percentage , of neutrophils ( Gr1+Ly6+ ) , was also reduced in the infected PND compared to WN mice ( Fig . 4F , right panel; p = 0 . 02 ) . Next we investigated whether malnutrition altered the localization of different cell populations in the lymph nodes before or after L . donovani infection , since this would influence cell and parasite trafficking to and within the LN . Staining of serial lymph node sections for B cells , T cells , endothelial cells , macrophages/monocytes , DCs and FRC did not reveal any difference in the distribution of these cells in the LNs of PND compared to WN mice ( Fig . 5A–D ) . In order to define the location of DCs , LN sections were co-stained for the DC marker , CD11c , and the FRC marker , ER-TR7 . CD11c+ DCs were distributed similarly in the paracortex of all the different groups of mice . CD11c+ DCs associated with conduit structures were probably LN resident DC while the DCs located free in the paracortex were likely DCs that had migrated from either the dermis or conduit system ( Fig . 5B ) . The CD11c marker does not distinguish between these cell populations . Similarly , LN sections stained for MOMA-2 and ER-TR7 showed a comparable staining pattern for macrophages/monocytes between PND and WN mice ( Fig . 5C ) . In addition , the number of B cell follicles was comparable between the two groups ( Fig . 5D ) and this was consistent with the result of H&E staining ( data not shown ) . To distinguish between LN resident DCs or dermal DCs and Langerhans cells that had migrated in response to L . donovani infection , we stained LN sections with antibody against CD205 [36] , [37] . Consistent with the result of CD11c staining , we could not detect any difference in the distribution of CD205+ in the lymph nodes of WN and PND infected mice ( Fig . 5E ) . The staining pattern of CD169+ cells was comparable between the PND and WN mice in the sub-capsular sinus region; however , there were more CD169+ cells in the cortical region of the WN infected mice ( Fig . 5F ) . Collectively , these data indicate that while malnutrition selectively reduced the number or percentage of FRCs and several myeloid cell populations in the LN ( resident and/or migratory DCs , MOMA2+ and CD169+ macrophages ) , it did not alter the localization of the FRCs or different phagocyte populations in the LN with the exception of reduced numbers of CD169+ macrophages in the sub-cortical region . The stromal network has multiple functional roles in controlling the immune response in lymphoid organs by influencing cell recruitment , migration , activation , and survival . The master player in this system is the FRC , which forms a network of conduits that allows the transport of small antigens from the subcapsular sinus directly to the subcortical T cell region and the delivery of cytokines and chemokines to the port of entry of lymphocytes from the circulation , the high endothelial venule ( HEV ) [25] , [38] . Apart from its participation in transporting the small molecules and antigens , the role of the conduit system in pathogen containment or dissemination had not been investigated . We hypothesized that malnutrition could alter the stromal architecture and conduit system to enable Leishmania-infected cells that have traversed the floor of the SCS to escape the LN and enter the blood stream through the HEV . To examine the influence of malnutrition on the integrity and the function of conduit system , WN and PND L . donovani-infected mice were injected subcutaneously with either high or low molecular weight ( MW ) fluorescently labeled dextran ( Fig . 6A ) . Under normal circumstances high MW dextran is not able to traverse the floor of the SCS , whereas low MW dextran was shown to readily cross the SCS and accumulate in the LN conduit network within a few minutes after cutaneous injection [39] . In infected WN mice , Texas Red-labeled low MW dextran was clearly observed to be co-localized with the lymph node conduits 3 minutes after injection . In the PND infected mice , a reduced quantity of low MW tracer was found co-localized with the LN conduit system ( Figs . 6B , 6C , 6F ) , but concomitantly a two-fold increase in low MW dextran accumulation was observed in the spleen ( Fig . 6D , 6F ) , suggesting that there was impaired retention of the low MW tracer in the lymph node conduit system of the PND mice . To further assess the function and integrity of the SCS/conduit network , we injected the L . donovani-infected PND and WN mice with high molecular weight dextran ( 500 and 2000 kD ) . Contrary to the findings with low MW dextran , the distribution of the high molecular weight dextran was limited to the subcapsular sinus ( Fig . 6E , 2000 kD shown ) and medullary sinuses with little accumulation in the cortex and without co-localization with the conduit network . The distribution and quantity of the high MW dextran in the LN were comparable between the PND and WN L . donovani-infected mice ( Fig . 6E , 6G ) indicating that there was no difference in transit of the fluorescent antigen from the skin to draining LN . Collectively , these data indicate that the SCS-conduit interface was intact in PND infected mice ( the high MW dextran did not gain an access to the conduit system ) , the integrity of the conduit was maintained without lateral leakage into the subcortical region , but the reduced retention of the small MW tracer led to greater accumulation in the spleens of PND mice . We investigated whether the reduced accumulation/retention of the low MW antigen in the conduit system of L . donovani-infected PND mice was accompanied by alteration of the architecture and the molecular components of conduit network . Staining of serial frozen sections of popliteal LNs of PND and WN infected mice with antibodies against reticular fiber components ( laminin , collagen IV , heparan sulfate proteoglycan , collagen I , collagen III , fibronectin , desmin and alpha smooth muscle actin ) , with or without co-staining of FRC ( ER-TR7 antibody ) , revealed no differences in the quantity or localization of any of these components in the lymph node of WN compared to PND infected mice ( Fig . 7A , 7D ) . To further assess the structure of the LN conduit system , we measured the length and the width of the reticular fibers in LN paraffin-embedded sections following reticulin stain . Consistent with the immunohistochemistry data , we did not find any significant difference in either the length or width of the reticular fibers from the two groups of mice ( Fig . 7B and 7E ) . Furthermore , investigation of the ultrastructure of the LN conduit system of PND and WN mice by transmission electron microscopy showed no differences in the structure of the conduit system basement membranes surrounding the collagen strands and other extracellular matrix components ( Fig . 7C ) . Together these data suggest that the alteration of the conduit system function in PND infected mice , which allowed small molecules to escape from lymph node conduit system to the spleen , was not the result of alteration of the gross structural framework of the stroma and conduit system . By inference , and together with the finding of reduced LN phagocytic cells , the escape of the low molecular weight dextran from the conduit is likely due to the reduced phagocytic capacity ( fewer DCs and macrophages ) associated with the conduits . To investigate whether there is a difference in the parasite localization within the lymph nodes of PND and WN mice , we infected mice with fluorescent-labeled L . donovani and examined its localization relative to FRCs ( ER-TR7 ) , macrophages ( MOMA-2 ) , DCs ( CD11c ) , and Langerhans cells ( CD205 and CD207 ) at 3 days post-infection and CD169+ cells at 2 hours post-infection . The pattern of cellular infection appeared to be similar between the PND and the WN infected mice . In both groups of mice , L . donovani could be observed in close association with the conduit system , but without complete co-localization with the FRCs ( Fig . 8A ) . There was , however , a high degree of co-localization of L . donovani with lymph node resident DCs , and to a lesser degree with macrophages , in the vicinity of the conduit system in both PND and WN mice ( Fig . 8B , E ) . The finding of the parasite in both groups of animals inside the high endothelial venule ( HEV ) ( Fig . 8A ) suggests that the parasite is able to transit through the conduit system , and it is likely that the reduced number of conduit-associated DCs in the PND mice enhance this transit to the systemic circulation . There was no obvious co-localization of the parasite with CD205+ cells ( Fig . 8C ) or CD207 ( data not shown ) Langerhans cells . There was co-localization of L . donovani with CD169+ cells in both PND and WN mice ( Fig . 8D ) . The co-localization of the parasite with lymph node resident DC together with the parasite association with FRC indicates that the parasite might go through the conduit system despite of the size exclusion properties of the collagen III core component of the conduit network , but this did not appear to be amplified in the PND host . We used flow cytometry to further quantify the level of infection of LN cell populations and did not observe any difference in the degree of infection of macrophages or FRC between the PND and the control mice ( Fig . 8H ) . However , there were fewer ( reduced total number but not percentage ) infected CD169+ macrophages and infected DCs in the PND mice ( p = 0 . 04 and p = 0 . 02 , respectively ) ( Fig . 8F and 8G ) , which probably can be attributed to the reduced total number of these cells in the PND mice ( see Fig . 4D ) .
Dysfunction of the immune system is the critical link in the vicious cycle of malnutrition and infection [40] , [41] . Our earlier work demonstrated that polynutrient ( protein , energy , zinc and iron ) deficiency led to increased dissemination of L . donovani from the skin to the spleen and liver , which was due to impaired barrier function and reduced parasite containment in the draining LN [23] . These studies utilized a murine model of malnutrition that mimicked the complex features of moderate childhood malnutrition found in resource-limited regions of the world , which typically involves deficiencies of protein and energy with superimposed deficits of other nutrients such as zinc [42] and iron [43] . In the work presented here , the malnutrition-related loss of LN barrier function with resulting early dissemination of the parasite was accompanied by reduced overall LN mass and cellularity , in particular reduced numbers of mononuclear phagocytes in the LN subcortical region ( many of which are associated with the conduit system ) and lining the floor of the subcapsular sinus . Furthermore , there were reduced numbers of FRC , which form the network of conduits that transport small molecules and antigens to the subcortical T cell regions , and there was evidence of altered conduit function . These data identify previously unrecognized effects of malnutrition on the LN and provide a foundation for understanding the early immunological events that lead to increased dissemination of L . donovani and perhaps other pathogens in the malnourished host . To investigate the mechanisms of early parasite dissemination in the polynutrient deficient mice , we used intradermal inoculation of metacyclic promastigotes to mimic the natural initiation of infection by delivery of infective stage of the parasites into the skin . Distinct from our previous work , we used an earlier parasite challenge ( one month after the initiation of the polynutrient deficient diet ) and lower inoculum size ( for more relevance to natural transmission ) coupled with a more sensitive assay to quantify the parasite burden . This approach has a limitation in that the inoculum lacked the sand fly salivary components that would be included with a natural inoculation and which have been shown to promote Leishmania infection [44] and could enhance the dissemination of parasites from skin to the viscera . Nevertheless , we found about a 16-fold reduction in the percent of lymph node barrier function , and conclude that the early parasite dissemination is the result of the impaired capacity of the lymph node to contain the parasite locally . Since the total extradermal parasite burdens ( local draining lymph nodes , spleen , and liver ) were comparable between the two groups , it appears unlikely that early increase in the parasite visceralization was due to defective local parasite killing in the lymph node ( although this is likely to be an issue later in the course of infection in the malnourished host [45] ) or a difference in the rate of the parasite multiplication . Furthermore , hematogenous dissemination of L . donovani from the site of skin infection is not likely to contribute significantly to the malnutrition-related parasite visceralization because the parasite burdens in the skin and draining LN were no different in the WN and PND mice at an early time point . While the effect of malnutrition on the LN has not been described previously , a number of malnutrition-related changes in the composition and structure of other lymphoid tissues have been reported , including ( 1 ) atrophy of thymus and spleen [8] , [32] , [46] , ( 2 ) reduced thymic cellularity attributed to enhanced thymocyte apoptosis and decreased intrathymic cell proliferation [47] , [48] , ( 3 ) alteration in the thymic microenvironment [49] , [50] , ( 4 ) reduced in vivo and in vitro bone marrow cell proliferation [51] , [52] , ( 5 ) loss of splenic lymphoid cells around the small blood vessels [8] , and ( 6 ) reduced number of splenic T lymphocyte subsets [53] . We did not find any remarkable difference in the gross structure or cellular distribution within the lymph nodes , however , consistent with the previous observations in the thymus and spleen we did observe a significant reduction in the weight and cellularity of the LN of the PND mice , whether they were uninfected or infected with L . donovani , when compared with their WN controls . Myeloid populations within the lymph node were most significantly affected by PND . Uninfected PND mice had fewer LN dendritic cells compared with the WN controls , but following infectious challenge reduction in LN dendritic cells , macrophages and neutrophils was evident . These findings , along with the 2-fold reduction in the number of the parasitized LN DCs , in the infected PND mice suggest that malnutrition contributes to parasite dissemination through several possible mechanisms . First , the reduced numbers of resident DCs and macrophages in the LN may lead to overwhelming of the phagocytic capacity of the organ with escape of parasites to the systemic circulation and visceral organs . The lower retention of the low molecular weight tracer within the lymph node conduit system and increased trafficking to the spleen in the PND infected , probably the result of reduced phagocytic capture , mice supports this possibility of increased dissemination of the parasite through the conduit system . Second , malnutrition may lead to altered migration and/or LN retention of parasitized DCs leading to increased parasite dissemination . In support of the later , it is commonly held that dendritic cells are the primary means by which Leishmania is transported from the site of skin infection to the lymph node [54] , and some studies have also implicated macrophage in this process [31] . We could not detect any co-localization of the parasite with CD205+ or CD207+ cells , which indicates that Langerhans cells do not play a role in moving the parasite to the draining lymph node . This is consistent with recent observations in another Leishmania infection model [54] . Altered DC migration and maturation , cytokine production , and adhesion molecule expression was demonstrated previously in human malnutrition [55] . This impaired DC function may be related to reduced leptin levels [56] , [57] , and/or increased levels of prostaglandin E2 [58] , both of which were found in earlier work to be abnormal in our model ( [23] , [59] , and GM Anstead , unpublished data ) , and therefore could play a role in altered migration of Leishmania-infected DCs in the PND host . The route through which infected DCs might disseminate is currently under investigation . Subcapsular sinus ( CD169+ ) macrophages , which line the floor of the subcapsular sinus and medulla of the lymph node and play a key role in the lymph filtration and the translocation of the large or particulate antigens across the sub capsular sinus lining to the cortex [60] , were reduced in the infected PND mice . A recent study showed that depletion of CD169+ cells led to dissemination of vesicular stomatitis virus through the lymphatics after subcutaneous inoculation of mice with the virus [35] . The reduced numbers of CD169+ macrophages in the infected PND mice may lead to impaired transmigration of the parasite to the LN cortical region and thus favor transit of parasites from the subcapsular sinus directly to the efferent lymph and dissemination to the bloodstream . The LN reticular network plays crucial functional and structural roles in the defense against pathogens by promoting interaction between T cells and antigen-presenting cells , enabling rapid transport of free antigens through the conduit system for uptake and presentation by resident DCs to T cells [25] , [61] , and helping in the recruitment , retention and proper localization of immune cells . FRCs establish the reticular network by secreting extracellular components to produce reticular fibers which are interweaved to form the conduit system [62] , [63] . It was reported that the FRC is a target cell during infection by multiple pathogens , particularly those that persist chronically , including L . major infection [34] . Our data showed that L . donovani was associated with the conduit , and was found co-localized with the resident DCs surrounding it , in both WN and PND mice . We did not identify infected FRC but found the numbers of FRCs were decreased significantly in the L . donovani infected PND mice . Since FRCs produce DC chemoattractants such as CCL19 and CCL21 [64] , the decreased number of FRCs may significantly alter chemoattraction and retention signals , possibly resulting in increased escape of parasite-loaded phagocytes from the lymph node to the visceral organs . Lymph flows from the afferent lymphatics into the lymph node subcapsular sinus , then to the conduit system and out through HEV into the bloodstream [39] . The presence of the parasite in association with the conduit system and in the HEV suggests that they may traverse the conduit system into the HEV to disseminate through the blood stream . Furthermore , the presence of the parasite in the lymph node very early in the infection suggests that the parasite may be carried through the lymph and enter subcortical region via the conduit network independent of migratory DCs . Evidence supporting this idea comes from the previous work that demonstrated activation of lymph node resident DCs surrounding the conduit system within a few hours of L . major inoculation , while migration of skin derived DCs to the LN was not evident until approximately 14 hours after infection [37] , [61] . Additionally , L . chagasi was found in the draining lymph node of infected hamster two hours after infection [65] . Since the LN conduit network allows only small molecules ( <70 kD ) to pass along the reticular fibers [25] , [39] , [66] , [67] , we suspected that there might be a breach in the integrity of the floor of the SCS allowing entry of the much larger parasites into the conduit system . However , we found the high molecular weight dextran was retained in the subcapsular sinus without association with the FRC network indicating that a functional barrier was intact . This suggests that parasite trafficking through the LN is an active process , perhaps mediated by transmigrating subcapsular sinus macrophages or migratory DCs , but it remains to be determined how the parasite escapes the size exclusion property of the reticular fiber to gain an access to the conduit system . The presence of comparable quantities of the high molecular weight dextran in the LNs of PND and WN mice indicates that the influx of the tracer from the skin to the LN was not altered in the PND mice and that the reduced amount of the low molecular weight dextran in the conduit system of PND mice was likely to be due altered transmigration and/or retention . In summary , to our knowledge this study is the first to describe the architecture and cellular composition of the lymph node in the malnourished host . Based on our findings , four possible scenarios could explain how malnutrition leads to the loss of lymph node barrier function and early dissemination of L . donovani . First , the reduced total number of DCs and macrophages ( in both the subcapsular sinus and subcortical regions ) , with the resulting decrease in numbers of parasitized cells in the lymph node of the PND mice , would translate to a reduction in overall phagocytic capacity of the lymph node as an organ and allow the escape of parasites . Second , the reduced number of CD169+ macrophages may lead to impaired parasite capture and transmigration of the infected phagocyte into the lymph node cortical region allowing the parasite to escape the lymph node through the efferent lymphatic to the bloodstream and the visceral organs . Third , the reduced number of LN DCs may also alter trafficking and/or reduced retention of parasitized DCs in the LN . Lastly , the altered function of the LN conduit system , which may be related to a deficiency in resident macrophages and DCs along the conduit system resulting in reduced capture of parasites as they transit through the conduit , could lead to increased dissemination through the HEV to the systemic circulation . While there is support for each of these scenarios from the data presented here , and they are not mutually exclusive , further work is warranted to clearly define the route and mechanisms of visceralization in the malnourished host . | The impact of malnutrition in the world is staggering . Malnutrition is thought to directly or indirectly contribute to more than half of all childhood deaths , most of them related to heightened susceptibility to infection . Visceral leishmaniasis ( VL ) , caused by the intracellular protozoan Leishmania donovani , is a progressive , potentially fatal infection found in many resource-poor regions of the world . Most people who get infected with this parasite have only an asymptomatic latent infection , however , people who are malnourished have a greatly increased risk of developing severe VL . We initiated these studies of an experimental model that mimics human childhood malnutrition to better understand how malnutrition increases the susceptibility to VL at the molecular and cellular level . In this model we found that malnutrition led to failure of the skin-draining lymph node to act as a barrier to dissemination . This loss of lymph node barrier function was associated with a significant reduction in the numbers of dendritic cells and macrophages , phagocytic cells that capture and kill invading pathogens , and alteration of the flow of lymph through the lymph node . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] | [
"immunopathology",
"immunity",
"immunology",
"host-pathogen",
"interaction",
"biology",
"microbiology",
"pathogenesis",
"parasitology",
"immune",
"system"
] | 2013 | The Malnutrition-Related Increase in Early Visceralization of Leishmania donovani Is Associated with a Reduced Number of Lymph Node Phagocytes and Altered Conduit System Flow |
Allostery is conformation regulation by propagating a signal from one site to another distal site . This study focuses on the long-range communication in DNA mismatch repair proteins MutS and its homologs where intramolecular signaling has to travel over 70 Å to couple lesion detection to ATPase activity and eventual downstream repair . Using dynamic network analysis based on extensive molecular dynamics simulations , multiple preserved communication pathways were identified that would allow such long-range signaling . The pathways appear to depend on the nucleotides bound to the ATPase domain as well as the type of DNA substrate consistent with previously proposed functional cycles of mismatch recognition and repair initiation by MutS and homologs . A mechanism is proposed where pathways are switched without major conformational rearrangements allowing for efficient long-range signaling and allostery .
Allostery is a fundamental part of many if not most biological processes . It is classically defined as the induced regulation at one site by an event at another distal site . Venerable models for allostery , such as the MWC ( Monod-Wyman-Changeux ) [1] and KNF ( Koshland-Nemethy-Filmer ) [2] models emphasize a mostly static picture of induced conformational changes . The MWC model proposes coupled conformational changes via a population shift while the KNF model highlights the induced-fit of a binding of a ligand via common communication routes . A broader view of allostery [3–6] emphasizes communication pathways via protein motions but without requiring actual conformational changes . The idea of this model is that relatively minor perturbations may shift communication between multiple pre-existing pathways . Such a mechanism has been demonstrated by nuclear magnetic resonance ( NMR ) experiments for the binding of cyclic-adenosine monophosphate ( cAMP ) to the dimeric catabolite activator protein ( CAP ) [7] as well as for allosteric regulation in Pin1[8] . Recent work based on Markov state models that integrate energetics and kinetics has added further nuances to the discussion by emphasizing both conformational and kinetic selection as the main mechanism of allostery in signaling proteins protein kinase A [9] and NtrC [10] . The idea of kinetic selection is consistent with a pathway selection mechanism without significant conformational changes . Recent reviews have attempted to integrate the different ideas into a unified view [11 , 12] with the main question being to what degree conformational dynamics plays a role . Likely , the degree of dynamics will depend on a given system and the economics of achieving allosteric signaling within the thermodynamic and functional constraints in the biological environment . One particular question that is central to this work is how long-range allostery can be achieved in very large systems where larger conformational changes and global selection mechanisms that are conceptually straightforward in smaller proteins could be more challenging to realize . It is difficult to obtain detailed insight into allostery from experiments , especially for larger and more complex systems , because NMR spectroscopy is generally limited to small and soluble proteins that can be easily labeled and expressed in large quantities . On the other hand , crystallography is not well-suited for studying allosteric effects due to its inherent dynamic nature . Computational approaches such as statistical coupling analysis ( SCA ) [13] , normal mode analysis ( NMA ) [14 , 15] , dynamical network analysis [16] , and Markov state model analysis based on extensive molecular dynamics simulations [9 , 10] offer complementary means for exploring allosteric mechanisms in biological systems . SCA , a bioinformatics-based method , obtains allosteric information by identifying coevolving residues from multiple sequence alignments , while NMA , a structure-based approach , suggests induced movements from a few robust low-frequency normal modes . Allosteric pathways obtained from these two methods would be encoded in the sequence and/or structure , but sensitivity to minor perturbations with this type of analysis is lacking . Dynamical network analysis [16] is based on molecular dynamics ( MD ) simulations and has been used to identify synchronous and/or asynchronous correlated residue motions in order to describe possible allosteric communication pathways . Examples of where this approach has been applied successfully to probe allosteric coupling include a tRNA-protein complex [16] , the M2 muscarinic receptor [17] , and cysteinyl tRNA synthetase [18] . Here , we used dynamical network analysis to develop a paradigm for allostery in very large multi-subunit complexes based on long-range signal propagating pathways in the MutS component of the methyl-directed DNA mismatch repair ( MMR ) system . MMR is responsible for correcting errors that escape immediate proofreading during DNA replication and the mechanism is widely conserved from prokaryotic to eukaryotic organisms . MMR alone can increase the accuracy of DNA replication by 20–400 fold [19] . While several components , such as MutS , MutL , MutH , nuclease , and polymerase , are needed to work together to complete DNA repair [20] , MutS is responsible for the initial recognition of DNA lesions , in particular mismatches and insertions or deletions ( IDLs ) . MutS is a homodimer , but , structurally and functionally , it acts as a heterodimer because only one subunit ( termed the ‘A’ chain in this paper ) directly contacts the lesion sites [21] . MutS homologs ( MSH ) in eukaryotes are heterodimers with differing substrate specificities . MutSα ( MSH2-MSH6 ) preferentially recognizes base pair mismatches and single base IDLs [22] , whereas MutSβ ( MSH2-MSH3 ) has a higher affinity and specificity for small DNA loops composed of 2–13 bases [23] . The crystal structures of prokaryotic MutS and its eukaryotic homologs , complexed with mismatched DNA heteroduplexes , feature a similar overall Θ shape [22 , 24–26] . Each subunit of MutS and MSH is comprised of five distinct domains ( see Fig 1 ) : the mismatch-binding domain ( MBD , domain I ) , the connector domain ( domain II ) , the lever domain ( domain III ) , the clamp domain ( domain IV ) , and the nucleotide binding domain ( ATPase , domain V ) [24] . The MBD and clamp domains interact with the bound DNA directly . The MBD contains a conserved , mismatch-identifying Phe-X-Glu motif , forming specific interactions with mismatches . The phenylalanine forms an aromatic ring stack on the 3′ side of the mismatched base [24 , 25] although there is also evidence for base flipping of the mismatched or neighboring base during the mismatch recognition process [27 , 28] . MSHβ , which specializes in the recognition of longer insertions/deletions , lacks this motif . The lever and connector domains connect the MBD and clamp domains to the ATPase domain . The ATPase domain is a conserved domain in the ABC ( ATP binding cassette ) superfamily . Biochemical studies have provided evidence that ATPase activities are coupled with DNA scanning , mismatch recognition , and repair initiation [29–31] . The different functional states are assumed to involve different conformations of MutS . The major states are MutS without DNA with open clamps , MutS scanning DNA in search of a mismatch with the clamps closed , MutS bound to a mismatch in the tightly DNA-bound conformation seen in crystallography , and a sliding clamp configuration where MutS is able to move away from the mismatch without scanning or complete dissociation from the DNA [32] . Based on the biochemical data , nucleotide binding and exchange to the ATPase domain appear to be key allosteric effectors coupled to DNA mismatch recognition that at least in part trigger changes between those functional states . This implies allosteric coupling between the mismatch binding site and the ATPase site over a distance of 70 Å [33] is essential for the biological function of MutS . Mismatched binding promotes exchange from ADP to ATP based on kinetic measurements of ATP hydrolysis [30 , 34 , 35] and results in asymmetric activity of the two ATPase domains [30] , whereas the sliding clamp state is supposed to be formed in ATP binding states [36–38] . Previous studies have examined MutS and eukaryotic analogs via molecular dynamics simulations [27 , 32 , 39–43] , but many mechanistic questions remain . Here , we subjected previously generated simulations of MutS , MutSα ( MSH2-MSH6 ) and MutSβ ( MSH2-MSH3 ) to dynamical network analysis to elucidate allosteric communication pathways between the structural domains in MutS and MSHs . In particular , we addressed the questions of how intra-molecular signaling could be accomplished over very long distances via protein dynamics and how small perturbations could affect the signal propagation . Previous work has suggested coupling between the MBD and ATPase domains , but mechanistic details and in particular the role of exchanging NTPs still remain largely unclear [43 , 44] . The dynamic network analysis applied here allowed us to probe for pathways connecting the domains in contact with the DNA and the ATPase domain . Furthermore , by comparing pathways in simulations with different nucleotides bound to MutS and different DNA substrates bound to MutSα and MutSβ we were able to develop hypotheses for how communication along those pathways may be shifted during the functional cycle of MutS and its homologs .
A number of very similar MutS and MSH crystal structures are available with different nucleotide bound states and different mismatches or IDLs . The structural variations that can be discerned primarily focus on the MBD , ATPase and clamp domains and involve mostly local side-chain displacements rather than larger conformational changes of the main chain . For example , the MutS crystal structures 1E3M ( with a single ADP ) [24] and 1W7A ( with bound ATP ) [33] differ by only 0 . 35 Å in the Cα coordinates after superposition . MD simulations paint a similar picture . In previous work from our group , MD simulations of MutS with all possible combinations of nucleotides bound to the ATPase dimer did not reveal large conformational changes of the overall MutS structure based on RMSD and clustering analysis [27] . A similar conclusion was found for Thermus aquaticus MutS in a recent study [44] , although different nucleotides bound to the ATPase domain were not examined . Taken together , this information has suggested that allosteric communication in this system likely takes place via subtle changes in local dynamics to achieve signaling in MutS rather than via conformational selection or induced conformational changes [3 , 44] . Average dynamical cross-correlation matrices ( DCCM ) were calculated from the MD simulations . Fig 1 compares the DCCMs between MutS simulations with different nucleotides . A comparison of the DCCMs after 50 , 100 , 150 , and 200 ns generally shows little change after 50 ns ( S1 Fig ) . This suggests that the correlations based on 200 ns trajectories are well converged consistent with a previous study [45] . In all cases , we found strong local correlation within domains but also weaker coupling between distant parts of the complex ( Fig 1 ) . Overall , different nucleotide bound states resulted in similar coupling patterns , but differences as a function of different nucleotide bound states can be discerned . For example , the positive MBD ( I ) -connector ( II ) A coupling is strongest in ADP-None , while the strongest positive MBD ( I ) -connector ( II ) B coupling is observed in None-ADP . Also in the case of ADP-None , the MBD and connector domains of subunit A are strongly negatively coupled with the lever and clamp domains of subunit B . The two clamp domains are positively coupled in cases of ATP-ADP and None-None , which are stronger than the others . The positive coupling between the two ATPase domains is strongest in ATP-ADP . Similar direct correlations between MutS domains have also been observed in other work based on MD simulations of Thermus aquaticus MutS [44] . However , while a direct correlation analysis suggests coupling , it does not provide complete information about the pathway ( s ) along which allosteric communication take place and it discounts the possibility of asynchronous communication via stochastic steps that would introduce a variable time delay between signal input and output along a given communication pathway . Next , we turned to dynamical network analysis to allow for a more dynamic model of allostery where direct correlations between distant sites are not required . In this approach , pathways connecting residue pairs along the shortest path with the highest pairwise local correlations based on the converged DCCMs from 200 ns MD sampling are determined . We focused our analysis on the functionally most relevant signal propagation between the MBD , ATPase , and clamp domains using specific key residues as anchor points ( S2 Table ) . A first set of pathways was determined between MutS-F36 , the key residue in direct contact at the mismatch site , and MutS-K620 , the key residue involved in binding the phosphate tails of NTPs in the ATPase domain . A second set of pathways was focused on the communication between the two ATPase domains connecting MutS-K620 in the A and B chains and a third set of pathways was constructed from MutS-K620 to MutS-N497 , which is the contact point of the clamp domain with the DNA opposite the mismatch site in the B subunit of MutS . Mapping of the resulting pathways onto the MutS structure is shown in Fig 2 . The computational analysis suggests multiple major pathways that vary as a function of the nucleotides bound to the ATPase domain . Within each major pathway , there are ensembles of similarly optimal minor pathways . The variability in the pathways was greatest within a given structural domain , where strong coupling between many residues allowed for many alternate , equivalent routes . However , connections between domains were limited to certain key residue pairs ( S3 Table ) that presented bottlenecks in the respective pathways . When employing network analysis to group strongly coupled residues into communities ( S2 Fig ) , these communication bottlenecks appear as critical inter-community edges that are hypothesized to correspond to switching points between major pathways when perturbed . Tables 1–3 quantify the features of the optimal pathways in terms of the number of steps ( hops ) required to traverse a path from the beginning to the end , a weight reflecting the degree of correlation along the optimal path , and the minimum pairwise correlation for any residue pair along the path . This analysis was carried out for each of the three sets of pathways as a function of different nucleotides bound in the ATPase domains . The algorithm employed here is designed to always find an optimal path connecting two given residues . In order to identify paths that are functionally relevant we focused on paths that stand out by having significantly lower weights than other paths while also requiring that the minimum correlation along the path was at least 0 . 7 . Our assumption is that even if a path has an overall low weight , it would not be an effective route of communication if it contained one or more links with poorly coupled residues . The overall premise of this study is the development of a dynamic allostery model for MutS since structural and previous simulation data suggest little conformational change as a function of nucleotides bound to the ATPase domains . Such a model implies the presence of communication paths between key structural elements ( MBD , ATPase , and clamp domains ) and the main result of this work is the identification of such paths in a nucleotide-dependent manner . Integrating previous biochemical data with such a dynamic allosteric model allowed us to arrive at the mechanism depicted in Fig 7 and described in more detail in the following: In the absence of DNA , the ADP-ADP state is presumed to be dominant [54] . The ADP-ADP state dissociates directly from DNA , while the binding of DNA induces the dissociation of one ADP molecule , more likely to be the one in B subunit . Therefore , the ADP-None state is presumed to be the mismatch scanning state [34 , 55] . Crystal structures of MutS were also obtained mostly in the ADP-None state [56–59] . Our results also support this idea . The MBDA strongly couples with the connector and ATPase domains in the ADP-None state via the proposed pathway ② that consists of a broad ensemble of individual paths with a few bottlenecks at domain boundaries . This communication is then proposed to result in exchange of ADP for ATP in the ATPaseA domain upon mismatch recognition [34 , 35] . In our model , the presence of ATP in the A site would abolish the communication between the MBD and ATPaseA domains because an optimal or suboptimal path via the connector domain is either absent altogether ( ATP-ADP ) or present with less favorable weights or low minimum pairwise correlations ( ATP-none , ATP-ATP in path ① ) that suggest inefficient coupling . At the same time , the ATPaseA -ATPaseB communication would engage the Walker B motif of chain B when ATP is present in the A site . We further hypothesize that ADP or ATP binding to the B site would follow , leading to the ATP-ADP state . Since the lifetime of the ATP-None state is believed to be short [54] this would occur quickly . Once the ATP-ADP state is reached , our model suggests that the ATPaseA domain connects to the lever instead of the connector . Because the connector primarily connects with the MBD domain while the lever domain provides a route to the clamp domain , we hypothesize that in the ATP-ADP state communication from the ATPaseA domain would be switched from the MBD domain to the clamp . At the same time , strong coupling between the ATPaseA -ATPaseB domains via the signature loop could disrupt the strong ATPaseB-clamp connection present in the ADP-None state and allow release of MutS from the mismatch site . The mechanism above postulates communication routes within the context of a dynamic allosteric mechanism that could be tested further experimentally , e . g . via mutations of pathway residues . Based on the dynamics sampled in the underlying simulations we are able to propose a structural basis for how pathways are switched in the presence of different nucleotides , however , the model is still lacking a clear mechanism for how ADP would be exchanged for ATP following mismatch recognition , for how the clamp domains would respond to signaling resulting from nucleotide exchange as proposed here , and what role MutL binding plays in this process . We speculate that the altered correlated dynamics induces subtle shifts in the overall conformational landscape which would favor ADP-ATP exchange and lead to clamp opening . To address this idea in more detail , significant additional simulations are required to probe the DNA binding process and clamp dynamics leading to the sliding clamp conformation in excess of the scope of the present work . Such a model could also conceptually integrate recent conformational landscape-based ideas of allostery with the communication-focused analysis presented here into a complete model for allostery in a large , complex system such as MutS where simpler concepts of conformational selection or induced-fit may not be able to adequately describe the allosteric mechanism . MutS can recognize a broad range of lesions , mismatches and IDLs , but MSHs have differentiated substrate specificities . MutSα ( MSH2-MSH6 ) primarily recognizes mismatches and single base IDLs , whereas MutSβ ( MSH2-MSH3 ) recognizes DNA loops composed of 2–13 bases . Based on previous simulations of MutSα and MutSβ with native and swapped substrates and no DNA at all [60] , we also analyzed how different DNA substrates would shift the signaling pathways identified via our computational analysis . In the MSH complexes we identified pathways analogous to paths ① , ② , and ③ in MutS ( see Fig 8 and details in S8 Fig and S9 Fig ) suggesting that the proposed communication pathways may be preserved in the eukaryotic homologs . There appears to be strong communication from the MBD through the connector domain when MutSα and MutSβ are bound to their native substrates ( a G:T mismatch and a four-nucleotide insertion loop ( IDL-4L ) , respectively ) . However , swapping the substrate would abolish that path in favor of coupling along the lever domain . Again , cancer-associated mutations in MSH6 and MSH2 map onto the paths , some at critical edges connecting different domains ( see S7 Fig and S8 Fig ) . Interestingly , communication between the MBD and ATPase domain of MSH3/MSH6 through the connector would also be present in the absence of DNA . These findings expand our allosteric model where effective communication between the MBD and ATPaseA domains ( and subsequent initiation of repair ) would depend on both the nature of the DNA substrate and the nucleotides bound in the ATPase domains .
Long-range signaling and allostery is a key mechanistic component of many large biomolecular complexes . Here , we present a detailed analysis of E . coli MutS and MSHs where several long-range signaling steps are essential for initiating DNA repair following mismatch recognition . Using dynamic network analysis based on extensive molecular dynamics simulations we developed a model consisting of a number of communication pathways that depend on strong local pairwise residue dynamical coupling where signaling would be expected to progress stochastically along those paths . In this model , different combinations of ATPase-bound nucleotides would result in switching between different pathways to implement the functional cycle of MutS without significant conformational rearrangements . A signaling mechanism based on pre-existing pathways that are switched on or off by different nucleotides and/or different DNA substrates is consistent with previous crystallographic and simulation studies that show surprisingly little structural variations in mismatch-bound MutS and homologs . The benefit of such a mechanism could be energetic economy , especially when considering the very long range over which the pathways appear to operate . Experimental validation of the hypotheses presented here could involve mutations of key residues , but it will also be interesting to see whether similar mechanisms are at play in other large enzymes . However , further computational studies will also be necessary to develop a more complete mechanistic understanding of how exactly signaling along the proposed pathways would promote and depend on nucleotide exchange and how it would lead to sliding clamp formation and complex formation with MutL .
MD simulations of the E . coli MutS protein bound to a G:T mismatch DNA ( PDB ID: 1W7A ) [33] were previously performed [27] . Each ATPase site may have three states: ATP , ADP or no nucleotide . All combinations of the three states in either of the two ATPase domains were simulated . They are denoted as ATP-None , None-ATP , ATP-ATP , ADP-None , None-ADP , ADP-ADP , ATP-ADP , ADP-ATP and None-None ( S1 Table ) . In this notation , the first nucleotide is present in the ATPase site of the mismatch-binding moiety ( subunit ‘A’ ) and the second one in the ATPase site of the non-mismatch-binding moiety ( subunit ‘B’ ) . Additional new simulations were carried out for five mutants of the E . coli MutS system to test the mechanistic hypotheses developed in this study: E169P ( ADP-None ) , L240D ( ADP-None ) , and Q626A ( ATP-ADP ) in chain A as well as L558R in either chain A or B ( ATP-ADP ) . These simulations were simulated using the same protocol as the previous simulations of the wild-type systems ( see below ) . MD simulations of human MutSα and MutSβ were started from the crystal structure 2O8B [22] and 3THX [26] ( MutSα/G:T and MutSβ/IDL-4L ) [61] . In MutSα and MutSβ structures , MSH6 and MSH3 are the mismatch-bound moieties ( equivalent to the A subunit in MutS ) , while MSH2 interacts with the DNA non-specifically ( equivalent to subunit B in MutS ) . Additional simulations were carried out for apo structures , where the DNA heteroduplex was removed ( MutSα/Apo and MutSβ/Apo ) , and for MutSα and MutSβ where the respective substrates were swapped ( MutSα/IDL-4L and MutSβ/G:T ) [61] . In total , 15 previous simulations and five new simulations ( S1 Table ) , each for at least 200 ns , were analyzed . All of the simulations were carried out with NAMD 2 . 8 [62] using the CHARMM27 force field [63] , the latest force field available at the time those simulations were initiated . All systems were solvated in explicit solvent using the TIP3P water model and sodium counterions to neutralize the systems . Simulations were carried out under periodic boundary conditions with the particle-mesh Ewald method [64] to calculate electrostatic interactions at constant temperature ( 300K ) and constant pressure ( 1 atm ) using a Langevin thermostat and barostat . The fully solvated systems consisted of about 165 , 000 atoms for the MutS systems and about 600 , 000 atoms for the larger MutSα and MutSβ systems . All of the systems remained overall stable with RMSD values of 3–5 Å for Cα atoms with respect to the initial experimental structures . Additional details of the system setup and simulation results are described in our previous papers [27 , 61] . VMD [65] was used to visualize and analyze simulations and generate structural figures . Allosteric networks within the proteins were identified using the NetworkView plugin of VMD [16 , 66] . The dynamic networks were constructed using data from our molecular dynamics simulations of the protein-DNA complexes described above , each sampled every 1 ps . For each molecular system , a network graph was constructed with two nodes for each nucleotide ( at N1/N9 and Pα/P ) , while protein residues were represented with a single node at the Cα position . All of the conformations from a given trajectory were pooled to calculate the local-contact matrix . A contact between two nodes ( excluding neighboring nodes ) was defined as within a distance of 4 . 5 Å for more than 75% of MD trajectories . The resulting contact matrix was then weighed by the correlation values of the two end nodes in the dynamical network as wij = −log ( |Cij| ) , where Cij are the elements of the correlation matrix calculated as Cij=⟨Δri⋅Δrj⟩/⟨Δri2⟩1/2⟨Δrj2⟩1/2 . The correlation matrices , also called dynamic cross-correlation matrices ( DCCM ) , were calculated using the carma software [67] . The length of a path is the sum of the edge weights between the consecutive nodes along this path . And the optimal ( shortest ) paths between two nodes in the network were obtained by the Floyd-Warshall algorithm [68] . The number of optimal paths that cross one edge is termed as betweenness of the edge . Suboptimal paths within a certain limit ( offset ) between the two nodes were also determined in addition to the optimal path . The number of suboptimal paths shows the path degeneracy . Communities were calculated based on the dynamical network by the Girvan–Newman algorithm [69] . The nodes in one community are more compactly interconnected than other nodes . All pathways were determined between residues in the MBD ( located within 10 Å of the mismatch site ) and residues in the ATPase domain ( located within 10 Å of a bound nucleotide ) or between residues in the clamp domain ( within 10 Å of DNA ) and residues in the ATPase domain ( within 10 Å of a bound nucleotide ) . The residue pairs with the shortest optimal path were finally selected as representative residues ( S2 Table ) . Suboptimal paths between specific residue pairs were calculated with edge length offsets of 3 , 5 and 10 for the MBD-ATPase , ATPase-ATPase , and ATPase-clamp interactions , respectively . | We are proposing a new model for how long-range allosteric communication may be accomplished via switching of pre-existing pathway as a result of only minor structural perturbations . The systems studied here are the bacterial mismatch repair enzyme MutS and its eukaryotic homologs where we identified strong communication pathways connecting distant functional domains . The functionally-related exchange of nucleotides in a distant ATPase domain appears to be able to switch between those pathways providing a new paradigm for how long-range allostery may be accomplished in large biomolecular assemblies . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"allosteric",
"regulation",
"enzymes",
"dna",
"clamps",
"enzymology",
"phosphatases",
"mutation",
"network",
"analysis",
"mismatch",
"repair",
"dna",
"dna",
"structure",
"enzyme",
"chemistry",
"computer",
"and",
"information",
"sciences",
"proteins",
"enzyme",
"regulation",
"biophysics",
"molecular",
"biology",
"adenosine",
"triphosphatase",
"physics",
"biochemistry",
"biochemical",
"simulations",
"point",
"mutation",
"nucleic",
"acids",
"genetics",
"biology",
"and",
"life",
"sciences",
"dna",
"repair",
"physical",
"sciences",
"computational",
"biology",
"macromolecular",
"structure",
"analysis",
"biophysical",
"simulations"
] | 2016 | Long-Range Signaling in MutS and MSH Homologs via Switching of Dynamic Communication Pathways |
The activating immunoreceptor NKG2D promotes elimination of infected or malignant cells by cytotoxic lymphocytes through engagement of stress-induced MHC class I-related ligands . The human cytomegalovirus ( HCMV ) -encoded immunoevasin UL16 subverts NKG2D-mediated immune responses by retaining a select group of diverse NKG2D ligands inside the cell . We report here the crystal structure of UL16 in complex with the NKG2D ligand MICB at 1 . 8 Å resolution , revealing the molecular basis for the promiscuous , but highly selective , binding of UL16 to unrelated NKG2D ligands . The immunoglobulin-like UL16 protein utilizes a three-stranded β-sheet to engage the α-helical surface of the MHC class I-like MICB platform domain . Intriguingly , residues at the center of this β-sheet mimic a central binding motif employed by the structurally unrelated C-type lectin-like NKG2D to facilitate engagement of diverse NKG2D ligands . Using surface plasmon resonance , we find that UL16 binds MICB , ULBP1 , and ULBP2 with similar affinities that lie in the nanomolar range ( 12–66 nM ) . The ability of UL16 to bind its ligands depends critically on the presence of a glutamine ( MICB ) or closely related glutamate ( ULBP1 and ULBP2 ) at position 169 . An arginine residue at this position however , as found for example in MICA or ULBP3 , would cause steric clashes with UL16 residues . The inability of UL16 to bind MICA and ULBP3 can therefore be attributed to single substitutions at key NKG2D ligand locations . This indicates that selective pressure exerted by viral immunoevasins such as UL16 contributed to the diversification of NKG2D ligands .
Human cytomegalovirus ( HCMV ) is a β-herpesvirus that causes lifelong asymptomatic infections in healthy individuals but endangers the lives of immunocompromised individuals and very young children [1] . Cytotoxic lymphocytes such as CD8 T cells and natural killer ( NK ) cells are essential for the control of HCMV infection [1]–[4] . HCMV possesses a broad arsenal of immune evasive strategies that counteract cellular immunosurveillance and ensure long-term persistence in infected human hosts [2] , [5]–[8] . One such strategy is the degradation of MHC class I molecules in order to subvert presentation of HCMV-derived peptide antigens to CD8 αβ T cells [2] , [6] , [8] . However , in line with the ‘missing-self-hypothesis’ , impaired MHC class I expression results in a decreased engagement of MHC class I-specific inhibitory NK cell surface receptors and thus may facilitate NK cell-mediated lysis of the infected cells [9] . NK cell activity , however , is not solely controlled by receptors that inhibit NK cell activation , but rather is determined by the integration of signals from both inhibitory and activating NK cell receptors [10]–[12] . A potent activating receptor that mediates NK surveillance of stressed cells such as infected or malignant cells ( ‘induced-self’ or ‘stressed-self’ recognition ) is NKG2D ( natural-killer group 2 , member D ) [11] , [13] , [14] . NKG2D is a C-type lectin-like homodimer expressed on NK cells and cytotoxic T cells [15] . In humans , NKG2D transmits activating ( NK cells ) or co-stimulatory ( CD8 αβ T cells and γδ T cells ) signals via the associated DAP10 adaptor [16] and is triggered through engagement of cell stress-inducible MHC class I-related ligands belonging to the diverse MIC ( MHC class I chain related molecule ) and ULBP ( UL16 binding protein ) families . Two MIC ( MICA and MICB ) and six ULBP proteins ( ULBP1-6 ) are currently known [2] , [13] , [14] , [17]–[19] . The ULBP proteins are also sometimes referred to as ‘retinoic acid early transcript’ proteins ( RAET; ULBP1/RAET1I , ULBP2/RAET1H , ULBP3/RAET1N , ULBP4/RAET1E , ULBP5/RAET1G and ULBP6/RAET1L ) . To thwart an antiviral NKG2D-mediated immune response , HCMV counteracts virally induced cellular expression of NKG2D ligands by means of several immunoevasins [2] , [5] , [8] . HCMV-encoded glycoproteins UL16 and UL142 selectively prevent the surface expression of MICB , ULBP1 and ULBP2 ( UL16 ) and MICA ( UL142 ) , respectively , through intracellular retention [2] , [20]–[23] . The significance of evasion from NKG2D-mediated immunosurveillance is further highlighted by the recent discovery that the HCMV gene UL112 is transcribed into a microRNA ( miRNA ) which specifically suppresses translation of MICB mRNA [24] . Although all NKG2D ligands share a MHC class I-like α1α2-platform domain [25]–[28] that binds NKG2D , UL16 does not bind to MICA , ULBP3 , ULBP4 or ULBP5 [18] , [29]–[31] . This selectivity is surprising since MICA and MICB are highly homologous in sequence ( 83% identical residues in the α1α2 region ) but much more distantly related to the ULBP molecules ( that share 21–29% identical residues in the α1α2 region with the MICs and 38–59% amino acid sequence identity among each other ) , which were originally discovered in a screen for UL16-binding proteins [18] , [32] . In order to elucidate the structural basis for the ability of UL16 to engage highly diverse NKG2D ligands and to compare this promiscuous binding mode to that of NKG2D , we determined the structure of the UL16 ectodomain in complex with the α1α2-platform domain of MICB ( MICBpf ) at 1 . 8 Å resolution ( Table 1 ) . We also expected structural insights into the selective UL16 binding to MICB ( but not MICA ) and to some ULBP family members as selective pressure exerted by viral immunoevasins such as UL16 may have contributed to the diversification of NKG2D ligands [2] , [17] , [23] . We find that UL16 , which possesses no structural homology to NKG2D , nevertheless employs a NKG2D-like binding mode to interact with MICB . Our results also offer structural explanations for the selective UL16 binding to some NKG2D ligands , and illustrate how the immunological arms race between a persistent pathogen and the human immune system may have driven the evolution of proteins of both , virus and host .
UL16 is a heavily glycosylated 50 kDa type I transmembrane glycoprotein whose structure could not be predicted from its primary sequence [33] . In order to obtain soluble and homogeneously glycosylated protein for our structural and surface plasmon resonance ( SPR ) studies , we expressed the UL16 ectodomain in Chinese hamster ovary ( CHO ) Lec 3 . 2 . 8 . 1 cells [34] . UL16 was co-crystallized with MICBpf refolded from E . coli inclusion bodies ( see Materials and Methods ) . The UL16 ectodomain folds into a modified version of the immunoglobulin ( Ig ) -like domain ( Figure 1A ) . The presence of nine β-strands , arranged in two antiparallel β-sheets ( formed by β-strands A , G , F , C , C′ , C″ and β-strands D , E , B , respectively ) and a central disulfide bond linking β-strands B and F clearly classifies it as a variable ( V-type ) Ig-like domain [35]–[37] . In contrast to classical V-type Ig domains , however , UL16 also has an additional N-terminal “plug” ( amino acids 27–50 ) , formed by a two-stranded antiparallel β-sheet ( β-strands X1 and X2 ) and a short 310-helix ( Figure 1A ) . The plug covers the concave side of the AGFCC′C″ β-sheet and is covalently linked to the Ig-like core with a disulfide bond between β-strands X2 and F . The UL16-MICBpf complex was partially deglycosylated prior to crystallization , leaving only single N-acetylglucosamine ( NAG ) molecules attached to glycosylation sites . While there is no evidence for O-linked glycosylation , our electron density maps provide clear evidence for the presence of NAGs at seven out of eight putative N-glycosylation sites ( asparagines 35 , 41 , 68 , 84 , 95 , 101 and 132 ) . Modeling experiments show that native glycosylation would effectively shield much of the UL16 surface from solvent ( Figure 1B ) . In particular , the outward-facing AGFCC′C″ β-sheet and the N-terminal plug are expected to be mostly covered with glycans in the fully glycosylated protein . By contrast , the solvent-exposed face of the DEB β-sheet is devoid of glycans and available for interactions with other proteins . The extracellular region of MICB consists of two structural domains , the α1α2-platform domain ( MICBpf ) and the C-type Ig-like α3-domain [25] . The α3-domain is present only in the MIC family members of NKG2D ligands , but not among members of the ULBP family [17] , [25]–[28] . Our SPR measurements ( Figure 2 and Table S1 ) yielded almost identical dissociation constants ( KD ) for the complexes formed by UL16 with MICBpf ( KD = 66 nM ) or the complete MICB ectodomain ( KD = 67 nM ) , respectively . Together with a previous report [29] , this demonstrates that the α3-domain does not contribute to UL16 binding . Based on these results , only MICBpf was expressed and used for co-crystallization with UL16 . As previously reported for the unliganded MICB [25] , MICBpf folds into a structure that closely resembles MHC class I molecules , with two long parallel α-helices , contributed by domains α1 and α2 , arranged above an eight-stranded antiparallel β-sheet ( Figure 1B; for nomenclature of domains and secondary structure elements see Figure 3A ) . Comparison of MICBpf with the structure of the unliganded MICB ectodomain [25] shows that the platform domain remains essentially unchanged upon engagement of UL16 ( root-mean-square deviation of 1 . 4 Å for 172 common Cα atoms ) . Although minor differences are seen within three surface-exposed loops and a short N-terminal helix ( α0 ) , the residues in these regions have elevated temperature factors and do not contact UL16 . UL16 primarily engages MICBpf via a predominantly hydrophobic , glycan-free ( see also Ref . [29] ) surface comprised of its DEB β-sheet and the adjacent β-strand A , with additional contacts provided by the DE-loop ( connecting β-strands D and E ) and four amino acids ( aa 160–163 ) at the C-terminus ( Figures 3B and 4 ) . This surface interacts with the two long parallel helices at the top of the MICB platform domain and the β5β6-loop connecting β-strands β5 and β6 of MICB ( Figures 3A and 4 ) , shielding an area of 2194 Å2 from solvent . With the exception of the MICB region that corresponds to the peptide-binding groove in MHC class I proteins , the contact area contains few interfacial solvent-filled cavities . The complex features good surface complementarity ( Sc = 0 . 77 ) and is highly curved ( planarity = 4 . 0 ) [38] , [39] . Its overall organization resembles a saddle with two stirrups ( UL16 ) that is mounted on horseback ( MICBpf ) ( Figure 4A , see also Figure 1B ) . The saddle is formed by the DEB β-sheet , whereas the stirrups are contributed by the DE-loop and the C-terminus on either side of the sheet . To facilitate discussion of interactions , we have divided the UL16-MICB interface into three regions ( A , B and C , Figure 4 ) . Contact region A , which is located at the center of the complex and mostly hydrophobic in nature , contributes 54% of the total contact area . Interactions predominantly involve residues within the DEB β-sheet and β-strand A of UL16 . Eight UL16 residues ( Trp54 , Leu56 , Met59 , Ile61 , Ile63 , Tyr125 , Leu110 and Leu114 ) define a compact hydrophobic face that interacts with non-polar regions of MICBpf residues in its central α3-helix . These interactions are augmented with a salt bridge between UL16 Asp112 and MICB Lys152 and a number of mostly water-mediated hydrogen bonds ( Figure 4B ) . Contact region B , with 23% of the total contact area , is located at one end of the DEB β-sheet and within the DE-loop of UL16 . UL16 residues in this region contact several acidic residues ( Glu64 , Asp65 and Glu68 ) in the α1-helix of MICBpf , mostly via polar interactions ( Figure 4C ) . Contact region C , which contributes 23% to the total contact area , is located on the other side of the UL16 saddle . Here , the C-terminus of the UL16 ectodomain interacts with the β5β6-loop and the N-terminus of MICBpf via a mixture of hydrophilic and hydrophobic contacts ( Figure 4D ) . The overall architecture of the complex , with its large contact area and substantial number of interactions between contacting residues , indicates tight binding , which is in agreement with our SPR data that place the affinity of UL16 for MICBpf at 66 nM ( Figure 2 and Table S1 ) . A crystal structure of the NKG2D homodimer bound to MICB is unavailable . However , the NKG2D structure in complex with the highly homologous MICA protein [26] shows that both NKG2D monomers make extensive contacts with the long helices at the top of the MICA α1α2-platform domain . The NKG2D-MICA complex buries a surface area of 2170 Å2 , which is almost exactly the same area buried in the UL16-MICBpf complex . A superimposition of the two complexes demonstrates that contacts formed by UL16 overlap substantially with those made by one NKG2D monomer ( Figures 5A , B ) . One could therefore envision a scenario in which UL16 acts as a direct competitor for NKG2D [18] , perhaps even displacing it from its ligands . While the higher affinity of UL16 for MICB and ULBP1 ( KD values of 66 and 12 nM , respectively ) ( Figures 2 and S1 and Table S1 ) compared with the respective affinities of NKG2D for the same ligands ( KD values of 0 . 8 and 1 µM , respectively ) [40] would support this scenario , most reports to date indicate that UL16 acts inside the cell and is therefore unlikely to compete with NKG2D for ligand binding [2] , [17] , [20] , [21] . The detailed comparison of the central contact regions in each case reveals that , despite having entirely different folds , NKG2D and UL16 use an almost identical pattern of amino acid side chains to engage their ligands ( Figure 5C , see also Figures 3B and S2 ) . In UL16 , this pattern includes the MICB-contacting residues Ile63 , Lys123 and Tyr125 , while NKG2D uses an identical pattern of residues , Ile182 , Lys197 , and Tyr199 , to form very similar contacts with MICA . Remarkably , although the three side chains are contributed by different structural elements in each case , their position in space overlaps closely ( Figure 5C ) . This is also true for two additional UL16 residues , Leu110 and Leu114 , which are hydrophobic in nature and overlap with chemically related NKG2D residues Met184 and Tyr152 ( Figure 5C ) . Together , the five residues constitute a predominantly hydrophobic binding motif that is common to NKG2D and UL16 ( Figures 5B , C ) , and that forms the center of the interaction with the MIC molecules . This central binding motif is augmented by additional contacts , such as those mediated by UL16 residue Tyr65 and NKG2D residue Ser195 , that perform similar functions in the UL16-MICBpf and NKG2D-MICA [26] complexes ( Figures 3B and 5C ) . Since all MICA and MICB residues contacted by this central binding motif are identical , and since the structures of MICA and MICB superimpose well in this region , we conclude that UL16 mimics a key structural motif of NKG2D with an entirely different fold in order to engage MICB . Furthermore , we consider it likely that the central binding motif of UL16 also plays an important role in the recognition of other NKG2D ligands , for which structures of complexes with UL16 are not yet available .
Bacterial and viral pathogens often interfere with cellular activities and immunosurveillance processes to enhance their survival and effectiveness [41] . This is typically achieved by virulence factors , which imitate the function of a host protein by mimicking its key structural features . In the majority of such cases , pathogens first hijack and then manipulate host genes to produce structurally homologous versions of host proteins [41]–[45] . Thus , virulence factors and host proteins are derived from the same origin and arise from divergent evolution . However , structural mimics can also be generated through convergent evolution . Although differing in evolutionary origin and three-dimensional structure , the virulence factors have in this case evolved to mimic key structural features of cellular proteins . Examples for the latter strategy , which can only be revealed through structural analyses , are still exceedingly rare and are limited to a small number of virulence factors [41] , [46] , [47] . The comparison of HCMV UL16 with human NKG2D , reveals a striking example of convergent evolution [41] . A set of five predominantly hydrophobic core residues on the UL16 surface precisely mimics a set of five equivalent residues in the central region of the interface used by the structurally unrelated immunoreceptor NKG2D to interact with its ligands . As this central binding motif represents only a portion of the total interface between NKG2D and its ligands ( Figure 5 ) , one may wonder why UL16 mimics just this particular structural motif of NKG2D . McFarland et al . reported that residues constituting this motif ( Tyr152 , Met184 and Tyr199 ) form the basis for the highly degenerate ligand recognition mode of NKG2D [40] , [48] . They proposed a “rigid adaptation” mechanism , in which a rigid binding site on NKG2D uses the same set of predominantly hydrophobic core residues to make diverse interactions with a series of chemically and structurally distinct ligand residues . As an example , Tyr199 and Tyr152 of NKG2D can accommodate residues as diverse as Ala , Met or Phe at ligand position 159 [40] ( Figures 3A and S2 ) . Mimicry of these core residues likely enables UL16 to employ this binding mechanism of NKG2D to contact a similar set of ligands . The “rigid adaptation” concept is furthermore supported by the finding that UL16 engages its ligands via a rigid β-sheet , which does not allow for much conformational flexibility . The ligand residues contacted by NKG2D and UL16 in MICA and MICB , respectively , are Asp65 , Thr155 , Ala159 , Ala162 , Asp163 and the hydrophobic portions of the Arg/His158 side chain ( Figures 3A and 5C and S2 ) [26] , [40] , [48] . Since NKG2D and UL16 both evolved the same central binding motif in order to contact this specific set of ligand residues , the latter likely represent binding hot spots in MICA and MICB [48] . Furthermore , these residues probably are also of major importance for interactions with ULBP molecules ( Figures 3A and S2 ) . We note for instance that ( 1 ) based on the “rigid adaptation” concept the amino acid at ligand position 159 can be quite variable in size and chemical nature , ( 2 ) Asp163 is conserved in all NKG2D ligands , and ( 3 ) alanine and glycine dominate at position 162 . Unlike NKG2D , UL16 engages only MICB , ULBP1 , ULBP2 and ULBP6 , but not MICA , ULBP3 , ULBP4 and ULBP5 [18]–[20] , [29]–[31] . Our SPR measurements show that UL16 binds MICB with high affinity , whereas the affinity of UL16 for MICA is negligible ( Table S1 ) , in line with earlier studies [2] , [21] , [29] . Given the high degree of similarity between MICA and MICB at the sequence and structural level , the inability of UL16 to engage MICA is puzzling . In order to better understand the structural parameters that guide UL16 binding to MICB vs . MICA , Spreu et al . [29] assayed binding of soluble UL16-Fc to MICB chimeras in which they had exchanged domains , subdomains and single amino acids of MICB against equivalent regions of MICA . These experiments clearly demonstrated that recognition by UL16 is linked to residues projecting from the helical structures in the MICB α2-domain . However , the molecular mechanism by which these residues confer selectivity remained unclear . The crystal structure of the UL16-MICB complex now allows us to identify the key determinants of NKG2D ligand binding to UL16 . Our structural alignment of MICA and MICB identifies only seven MICB residues that contact UL16 in the complex and that are replaced by other amino acids in MICA ( Figure 3A ) . Residues at positions 64 , 71 , 75 , 102 and 158 can assume alternate conformations that would not interfere with binding , and could in some cases even mediate favorable contacts with UL16 . Therefore , their effect on UL16 binding is likely to be negligible ( see also Ref . [29] ) . Replacement of α1-domain Glu68 with glycine ( Figure 4C ) in MICA would eliminate several hydrophobic contacts and three hydrogen bonds with UL16 residues 117 and 118 , and could therefore conceivably have a negative effect on UL16 binding . However , as complete replacement of the α1-domain of MICB by MICA ( including residue Glu68 ) did not significantly affect UL16 binding [29] , residue 68 is probably not a key determinant of UL16 binding . On the other hand , however , Gln169 in the α2-domain of MICB is likely to be critical . Our structure shows that substitution of Gln169 with arginine , which is present at this position in MICA , would lead to steric clashes with UL16 residues Met59 and Leu161 ( Figure 6A ) that would prevent binding . This is in perfect agreement with previous experiments demonstrating that MICB carrying a Gln169Arg substitution no longer bound UL16 [29] . We consider it in fact likely that the side chain at position 169 is not only the key determinant of selective UL16 binding to the MIC molecules but all NKG2D ligands , which is based on the following reasons . ( 1 ) All NKG2D ligands that carry a glutamine or glutamate at position 169 , i . e . MICB , ULBP1 , ULBP2 and ULBP6 , bind UL16 , while all ligands that have an arginine at this position , i . e . MICA , ULBP3 and ULBP4 , do not bind UL16 ( Figure 3A ) . Although ULBP5 also carries a glutamate at position 169 and should therefore bind UL16 , Wittenbrink et al . demonstrated by mutational studies that a substitution in the α2-domain , which is unique among all NKG2D ligands ( Figure 3A ) , prevents binding of ULBP5 to UL16 [31] . ( 2 ) Arg169 has a similar conformation , stabilized by contacts with surrounding hydrophobic residues , in the unliganded [27] and liganded [26] MICA structures ( Figure 6A ) . In this orientation , however , the Arg169 side chain would clash with UL16 residues . Modeling suggests that the arginine side chain could adopt only a single rotamer conformation , sandwiched between the hydrophobic side chain regions of Leu172 and Lys173 , that would not result in steric clashes with UL16 ( Figure 6A ) . However , such a rotamer is only seen in 2% of all observed arginines [49] . ( 3 ) The conformation of Arg169 in the ULBP3 structure [28] , which is held in place by a salt bridge to Asp170 , would also clash with UL16 ( Figure 6A ) . A similar arrangement of Arg169 can be expected for ULBP4 , where Asp170 is replaced with glutamate ( Figure 3A ) . We note that Arg169 is not located near the NKG2D binding site and therefore does not play a role in the interaction of either MICA or ULBP3 with NKG2D . A second important requirement for binding of NKG2D ligands to UL16 is the presence of a small hydrophobic side chain at position 162 . In the UL16-MICBpf complex , Ala162 faces towards Tyr125 , a UL16 footprint residue ( Figures 5C and 6B ) . With the exception of ULBP3 , which has an arginine at this position , all other NKG2D ligands have either an alanine or a glycine at position 162 ( Figures 3A and S2 ) . The long and positively charged Arg side chain of ULBP3 would clash with several UL16 residues ( Figure 6B ) , likely contributing to the failure of UL16 to bind ULBP3 [2] , [17] , [18] , [21] ( Table S1 ) . Interestingly , Arg162 would also clash with Met184 of the L2-loop of NKG2D in its MICA-liganded form . To allow for ULBP3 binding , NKG2D undergoes a conformational adjustment in which the L2-loop displaces Met184 , resulting in sufficient space for the accommodation of Arg162 ( Figure 6B ) . However , the rigid DEB β-sheet of UL16 , which would not allow for such larger conformational adjustments , is unlikely to accommodate Arg162 . Taken together , these analyses suggest that some NKG2D ligands apparently bypass intracellular retention by UL16 through alteration of a small number of key residues at strategic locations of their potential UL16 binding interface . We therefore consider it likely that the selective pressure exerted by UL16 contributed to drive the diversification of NKG2D ligands , which eventually may have led to the emergence of non-UL16 binding ligands such as MICA and ULBP3 [2] , [5] , [17] , [18] , [23] . Further support for an HCMV-driven diversification of NKG2D ligands comes from studies by Cosman and colleagues showing that the HCMV immunoevasin UL142 targets most MICA allelic variants except MICA*08 [2] , [5] , [22] . Intriguingly , MICA*08 contains a truncated cytoplasmic domain and is by far the most frequent MICA variant in many populations [22] . As yet , no direct interaction of UL142 and MICA has been shown and the molecular mechanisms of MICA sequestration by UL142 are unknown . In contrast to UL16 , UL142 and MCMV-encoded immunoevasins m145 , m152 , and m155 that suppress surface expression of mouse NKG2D ligands MULT-1 , RAE-1 , and H60 , respectively , are predicted to have an MHC class I-like fold [2] , [22] , [23] , [50]–[54] . It will be of great interest to determine the structural basis of NKG2D ligand engagement by MHC class I-like HCMV immunoevasins and to compare these interactions of two MHC class I-like molecules to those of the NKG2D-like ligand binding mode of UL16 . NK receptors binding to MHC class I or class I-like molecules belong to two structurally distinct families , the Ig superfamily and the C-type lectin superfamily [10] . While NKG2D belongs to the latter group , our structural analysis shows that UL16 assumes an Ig-like fold . Therefore , one may ask whether UL16 is related to the Ig-like NK receptors that bind MHC class I molecules , such as the leukocyte Ig-like receptors ( LIRs ) or the killer immunoglobulin-like receptors ( KIRs ) . Structures of LIR-1 in complex with HLA-A2 [10] and with the HCMV MHC class I decoy UL18 [45] show that , in both cases , LIR-1 contacts β2-microglobulin and the α3-domain of the HLA-A2 and UL18 ligands via loops located at the interdomain hinge region of its two tandem Ig domains . In contrast KIRs , like UL16 , engage the α-helical parts of the platform domain of MHC class I molecules , but , similar to LIRs , employ loops at the interdomain hinge region of their Ig domains for this interaction [10] . Therefore , LIRs and KIRs exhibit an MHC class I-binding mode that is distinct from that used by UL16 . Since there is also no obvious sequence homology between these Ig-like NK receptors and UL16 , we favor the view that UL16 evolved independently , mimicking a central binding motif of the structurally unrelated NKG2D immunoreceptor . To the best of our knowledge , the structure presented here is the first structure of a viral immunoevasin in complex with a stimulatory NK receptor ligand as well as the first reported case of structural mimicry through convergent evolution of a human immunoreceptor by a viral immunoevasin . The results of our structural analyses revealed that HCMV and humans independently evolved two structurally distinct receptors , NKG2D and UL16 , that share the same central ligand binding motif in order to achieve promiscuous binding to MIC and ULBP molecules . Our findings provide new insights into the structural basis of the evolutionary struggle between persistent viruses and cellular immune surveillance , exemplified by the promiscuous binding mode of the HCMV immunoevasin UL16 and the diversification of NKG2D ligands .
All SPR experiments were performed and evaluated as described previously [55] . Using two consecutive flow cells on a CM5 biosensor chip , MICBα1α2 ( MICBpf ) and MICBα1α3 ligands , respectively , were each covalently immobilized on the surface of the downstream ( experimental ) flow cell via amine-coupling chemistry ( GE Healthcare ) following manufacturer's instructions , while the surface of the upstream ( reference ) flow cell was subjected to the same coupling reaction in the absence of protein . For the Protein A-G chip preparation , an amount of 3500 RU ( resonance units ) of recombinant Protein A-G ( BioVision ) was covalently immobilized to the upstream ( reference ) and downstream ( experimental ) flow cells of a CM5 biosensor chip ( GE Healthcare ) by amine-coupling chemistry ( GE Healthcare ) . Fc-tagged ULBP1 , ULBP2 , ULBP3 ( all R&D Systems ) , ULBP4 and ULBP5 ligands [31] were diluted in HBS-EP ( 10 mM HEPES , 150 mM NaCl , 3 mM EDTA , 0 . 005% ( v/v ) Surfactant P20 , pH 7 . 4 at 25°C ) and noncovalenty bound to the experimental flow cell surface . In all experiments , untagged , monomeric UL16 analyte was serially diluted in running buffer and injected in series over the reference- and experimental biosensor surface at 50 µl/min . After each cycle using a Protein A-G chip , the biosensor surface was regenerated ( stripped of any remaining analyte and ligand ) with two 1 min injections of 10 mM glycine pH 1 . 7 . CM5 chips were not regenerated . For crystallization , complex at 15 mg/ml was mixed in a 1∶1 ratio with a reservoir solution containing 0 . 2 M ammonium sulfate , 0 . 1 M sodium cacodylate pH 6 . 5 , and 25% PEG 8000 . Crystals grew at 4°C over a time period of 4 months using the hanging drop vapor diffusion method . They were soaked in reservoir solution enriched with 15% ethylene glycol , and then flash frozen in liquid nitrogen prior to data collection . The crystals belong to space group P212121 and contain two complexes in the asymmetric unit . All diffraction data were collected at 100 K and a wavelength of 1 . 0013 Å at the Swiss Light Source ( SLS , Villigen , Switzerland ) beamline X06SA using the PILATUS 6M detector . Data were indexed , integrated and scaled with XDS [56] , and the structure was solved by molecular replacement as implemented in PHASER [57] using the unliganded MICB structure [25] ( PDB code 1JE6 ) as search model . The initial density map already clearly showed the approximate location of the UL16 molecules . Phases were then improved through non-crystallographic symmetry averaging using RESOLVE [58] . Structural refinement was performed with PHENIX [59] and model building was done with Coot [49] . Refinement included TLS-refinement of 26 TLS groups assigned by the TLSMD Server [60] . A data set for Rfree calculation was generated with 5% randomly selected reflections , and refinement progress was monitored by the decrease of R and Rfree throughout . The final model has R and Rfree values of 17 . 7% and 21 . 5% , respectively , and was validated using PROCHECK [57] and WHAT_IF [61] . Secondary structure elements were assigned with DSSP [62] . Structural figures were created with PyMOL [63] . Eight potential N-linked glycosylation sites were identified in the UL16 ectodomain . Six of the possible eight asparagine residues ( Asn41 , 68 , 84 , 95 , 101 , 132 ) carry NAG residues that are clearly defined by electron density ( Figure 1A ) . While extra density is present at the seventh residue , Asn35 , this density is not well defined , and no NAG residue was built at this location . No extra electron density is observed at the final asparagine , Asn145 , and thus this residue is either not glycosylated or carries an especially flexible glycan moiety . We note that Asn145 is close in space to Asn35 , which is glycosylated . In order to produce a realistic estimate of size and distribution of the glycan structure of native UL16 ( Figure 1B ) we used the GlyProt [64] online server and modeled hybrid and complex glycans linked to the seven Asn residues with NAG electron density . Atomic coordinates and structure factors have been deposited with the Protein Data Bank under accession code 2wy3 . | Cytotoxic lymphocytes such as natural killer ( NK ) cells or CD8 T cells have the ability to detect and destroy cells infected by viruses . They therefore are tools on which the human immune system critically depends in order to control viral infections . To avoid discovery by cytotoxic lymphocytes and to allow for longtime persistence in the human host , the human cytomegalovirus ( HCMV ) has developed a multitude of immune evasive strategies that are mediated by so-called immunoevasins . We present here a structure-function analysis of one of the best-known HCMV immunevasins , UL16 , and its interaction with a cellular ligand for NK cells , MICB . The normal function of MICB is to activate NK cells by engaging the most well-known NK receptor , NKG2D . Our results provide molecular evidence for the strategy used by UL16 to disable NK cell activation . In a rare example of structural mimicry that has likely arisen through convergent evolution , UL16 mimics a central binding motif of the structurally unrelated NKG2D protein . This allows UL16 to engage and disable several diverse NKG2D ligands , while others have apparently evolved to escape recognition by UL16 through alteration of key residues at strategic locations . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"biochemistry/molecular",
"evolution",
"virology/virulence",
"factors",
"and",
"mechanisms",
"virology/virus",
"evolution",
"and",
"symbiosis",
"immunology/immunomodulation",
"immunology/immune",
"response",
"immunology/innate",
"immunity",
"infectious",
"diseases/viral",
"infections",
"biophysics/biomacromolecule-ligand",
"interactions",
"immunology/immunity",
"to",
"infections"
] | 2010 | Structure of the HCMV UL16-MICB Complex Elucidates Select Binding of a Viral Immunoevasin to Diverse NKG2D Ligands |
TRIM5α is a restriction factor that limits infection of human cells by so-called N- but not B- or NB-tropic strains of murine leukemia virus ( MLV ) . Here , we performed a mutation-based functional analysis of TRIM5α-mediated MLV restriction . Our results reveal that changes at tyrosine336 of human TRIM5α , within the variable region 1 of its C-terminal PRYSPRY domain , can expand its activity to B-MLV and to the NB-tropic Moloney MLV . Conversely , we demonstrate that the escape of MLV from restriction by wild-type or mutant forms of huTRIM5α can be achieved through interdependent changes at positions 82 , 109 , 110 , and 117 of the viral capsid . Together , our results support a model in which TRIM5α-mediated retroviral restriction results from the direct binding of the antiviral PRYSPRY domain to the viral capsid , and can be prevented by interferences exerted by critical residues on either one of these two partners .
Retroelements constitute important evolutionary forces for the genome of higher organisms , yet their uncontrolled spread , whether from endogenous loci or within the context of retroviral infections , can cause diseases such as cancer , autoimmunity and immunodeficiency , including AIDS . Correspondingly , a variety of host-encoded activities limit this process , behaving as the arms of a line of defence commonly called intrinsic immunity , which notably contributes to restricting the cross-species transmission of retroviruses [1] . The product of the Friend virus susceptibility 1 ( Fv1 ) gene , which shares similarity with the gag region of an endogenous retrovirus , conditions the susceptibility of various mouse strains to murine leukemia virus ( MLV ) [2 , 3] . N-tropic and B-tropic MLV strains replicate in Swiss/NIH and in Balb/c mice , respectively , reflecting the presence of either the n or the b allele of Fv1 in the genome of these animals . A critical determinant of the differential sensitivity of MLV strains to Fv1 lies in amino acid 110 of the viral capsid ( CA ) , which is an arginine in the prototypic N-tropic MLV and a glutamate in its B-tropic counterpart [4 , 5] . Moloney MLV ( Mo-MLV ) harbors an alanine at this position and escapes both Fv1n and Fv1b , hence is termed NB-tropic . N-MLV is restricted too in some non-murine mammalian cells , including of human origin , which do not encode Fv1 . Blockade is in these cases mediated by TRIM5α , a member of the tripartite motif ( TRIM ) family of proteins [6–10] . TRIM5α also prevents the cross-species transmission of primate lentiviruses . The orthologues present in Old World monkeys , including macaque rhesus , restrict human immunodeficiency virus type 1 ( HIV-1 ) and N-MLV , while those from New World monkeys , tend to restrict simian immunodeficiency virus of macaques ( SIVmac ) and for some species N-MLV but not HIV-1 [11–13] . Human TRIM5α ( huTRIM5α ) blocks N-MLV , but is only weakly active against SIVmac and HIV-1 [6–9 , 11 , 14] . All TRIM proteins contain a RING , a B-Box and a coiled-coil region , which together constitute the so-called RBCC domain [10 , 15 , 16] . TRIM5α further harbors at its C-terminus the PRYSPRY or B30 . 2 domain , responsible for the viral capsid-specific capture of restricted viruses [17–19] . Sequence alignments of the PRYSPRY domains of various primate TRIM5α and other related TRIM proteins reveal 4 variable regions ( V1 , V2 , V3 and V4 ) , predicted to constitute surface-exposed loops based on the structure of the homologous domain of related proteins [12 , 20–23] . While V1 , V2 and V3 were all found to contribute to the antiviral specificity of TRIM5α orthologues [11] , V1 was shown to play a most critical role in this process . Within this loop , removing a positive charge at position 332 or substituting residues 335 to 340 by an eight amino acid rhesus sequence confers huTRIM5α with the ability to restrict HIV-1 [11 , 24 , 25] . Conversely , introducing residues 335–340 of huTRIM5α at the corresponding locus of rhesus monkey TRIM5α ( rhTRIM5α ) enhances the N-MLV blocking activity of the simian protein [26] . The present study was designed to define further how TRIM5α recognizes retroviral capsids . Our results indicate that , if huTRIM5α efficiently restricts only N-MLV and not B-MLV or Mo-MLV , this is due to the negative influence of a key residue in V1 . Conversely , MLV is capable of avoiding restriction via the interdependent influences of a cluster of amino acids exposed at the surface of its capsid .
In order to characterize the interaction between huTRIM5α and its viral targets , we introduced amino acid changes in the central V1 region of its PRYSPRY domain . Cell lines stably expressing the resulting mutants were generated by retroviral vector-mediated transduction of permissive Fv1-null Mus dunni tail fibroblasts ( MDTFs ) ( Figure 1 ) . A first series of MDTF derivatives expressing huTRIM5α mutants carrying single alanine substitutions at positions 334 to 339 were challenged with MLV- or HIV-derived green fluorescent protein ( GFP ) -expressing vectors , scoring infection by fluorescence-activated cell sorting ( FACS ) analysis ( Figures 1A and 2A ) . All mutants conserved the ability to restrict N-MLV , albeit at a slightly reduced efficiency for some ( e . g . , F339A ) . None could block HIV-1 , except for F339A that was lowly active . In contrast , replacement of tyrosine336 by alanine yielded a mutant capable of efficiently blocking B-MLV and , to a small extent , Mo-MLV . We examined the step of the B-MLV replicative cycle targeted by this expanded-spectrum huTRIM5α mutant . Several reports have demonstrated that N-MLV blockade by huTRIM5α occurs at an early post-entry stage , before reverse transcription [7 , 13 , 27] . In contrast , the only restriction activity so far identified against B-MLV is mediated by Fv1n , which allows viral DNA synthesis to proceed but inhibits viral nuclear import [28 , 29] . We thus infected MDTF cell lines expressing either wild-type or Y336A huTRIM5α , or control cells , with equal doses of N- , B- or Mo-MLV vectors and monitored the accumulation of reverse transcription products by PCR , using primers that amplified elongated minus-strand DNA ( Figure 2B ) . All three vectors yielded readily detectable reverse transcripts in control cells at 6 hours post-infection . Consistent with previous studies , N-MLV DNA levels were significantly reduced in the presence of wild-type huTRIM5α , whereas B-MLV escaped this effect . In contrast , both N- and B-MLV exhibited strikingly reduced amounts of reverse transcripts in cells expressing huTRIM5αY336A . With Mo-MLV , a slight decrease in viral DNA was noted in cells expressing this mutant at 8 hours post-infection , compared with control or wild-type huTRIM5α-expressing cells . This inhibition was more obvious when the analysis was repeated at 5 days post-infection , that is , upon scoring the ultimate proviral load of the cells , which correlated with the results of the FACS analyses performed at the same time ( Figure 2B and 2C ) . Altogether , these data indicate that wild-type and huTRIM5αY336A similarly act before the completion of reverse transcription . To explore further the modalities by which the Y336A mutation renders huTRIM5α active against B-MLV , we generated MDTF cell lines expressing huTRIM5α derivatives with other amino acid substitutions or with a deletion at this position ( Figure 1B ) . N-MLV restriction was unaffected by any of these changes . In contrast , variable degrees of B-MLV restriction were observed ( Figure 3 ) . Introduction of other small amino acids besides alanine ( threonine , serine , cysteine and , to a lesser extent , proline ) , or removal of tyrosine336 , also conferred a gain-of-function phenotype to huTRIM5α . One mutant , huTRIM5αY336K , even acquired the ability to block Mo-MLV with a good efficiency ( Figure 3 ) . This did not simply reflect the presence of a positive charge at this position , because the Y336R mutation rendered huTRIM5α only weakly active against B-MLV and even less so against Mo-MLV , even though it might be due to its poor expression level . Notably , replacing tyrosine336 with glutamate or phenylalanine did not broaden huTRIM5α restriction beyond N-MLV . Moreover , none of the mutants acquired the ability to block HIV-1 . It was previously demonstrated that a single amino acid substitution or deletion at position 332 of huTRIM5α provides this molecule with the ability to block HIV-1 infection [24 , 25 , 30] . We thus asked whether a molecule carrying changes at both positions 332 and 336 restricts not only N- and B-MLV , but also HIV-1 . For this , we generated MDTF cell lines expressing a series of huTRIM5α double mutants , using various combinations of substitutions and deletions previously noted as broadening the spectrum of activity of the protein towards either B-MLV ( this work ) or HIV-1 [24 , 25 , 30] ( Figure 4A ) . Double mutants that included an alanine substitution at position 336 were either poorly expressed ( R332H/Y336A , data not shown ) or expressed but without antiviral activity even against N-MLV ( R332A/Y336A , Figure 4B and 4C ) . All other double mutants were stably expressed , and exhibited the MLV restriction patterns expected from the residue present at position 336 ( Figure 4B and 4C ) . Single mutants with arginine to proline , histidine , alanine or a deletion at position 332 significantly restricted HIV-1 , albeit it less efficiently than rhTRIM5α . However , none of the double mutants was effective against this virus . Furthermore , substitutions at position 332 somewhat reduced the ability of huTRIM5αY336K to restrict Mo-MLV ( Figure 4C ) . It thus appears that changes underlying the acquisition of B- and Mo-MLV restriction ability preclude further extension of the activity of huTRIM5α to HIV-1 . The effective blockade of both N- and B-MLV by several huTRIM5α mutants strongly suggested that the “canonical” amino acid 110 of the MLV capsid , the importance of which for Fv1n and Fv1b sensitivity has been extensively demonstrated [4 , 5 , 27] , did not play an essential role in the case of TRIM5α . The N- and B-MLV packaging constructs sequence used in the present study encode for viral capsids that differ in only three residues ( CA109 , CA110 and CA159 ) [27 , 31 , 32] . Exchanging the residue present in either virus at position 110 , through reciprocal arginine-glutamate exchanges , sufficed to confer wild-type huTRIM5α susceptibility or resistance to N- and B-MLV . However , both mutants were potently blocked by huTRIM5αY336A , which also restricted either N- or B-MLV modified to contain an alanine at this position ( Figure 5A ) . These data confirmed that TRIM5α can be altered to act in a CA110-independent manner . However , in the context of Mo-MLV , this capsid residue plays a pivotal role . Mo-MLV could indeed escape all forms of TRIM5α-mediated blockade when alanine110 of its capsid was changed to glutamate . Inversely , when arginine was introduced instead , Mo-MLV restriction by huTRIM5αY336K and huTRIM5αY336A was strengthened , and the virus became slightly sensitive to wild-type TRIM5α , as recently noted [33] ( Figure 5B ) . We then sought to define which other capsid residues influence MLV susceptibility to huTRIM5α-mediated blockade . For this , we focused on amino acids in N- or B-MLV that differ from Mo-MLV , and on positions previously demonstrated to influence restriction by huTRIM5α and/or Fv1 ( Figure 6A ) [34 , 35] . We found that single point mutations at positions 82 ( N to D ) or 117 ( L to H ) of capsid allowed B-MLV to escape completely huTRIM5αY336A restriction ( Figure 6B ) . Nevertheless , the influence of these two mutations was context-dependent , because when introduced in N-MLV they relieved neither wild-type nor huTRIMaY336A-mediated restriction ( Figure 6B ) . Furthermore , the newly introduced residues are those naturally present in Mo-MLV , which is blocked by huTRIM5αY336K ( Figures 3 and 5B ) . The testing of a high number of additional single and combined mutants confirmed that MLV susceptibility to TRIM5α-mediated restriction is dictated by the interdependent influences of capsid residues 82 , 109 , 110 and 117 with a minor modulation by residue 159 ( Figure 7 ) .
It is suspected , albeit not yet formally demonstrated , that TRIM5α-mediated retroviral restriction proceeds through the direct binding of the antiviral PRYSPRY domain to the capsid of incoming viruses [13 , 17 , 18 , 24 , 27] . The present study , which demonstrates that the consequences of mutations in the huTRIM5α PRYSPRY V1 can be counterbalanced by changes in the MLV capsid , and vice versa , lends strong credence to such a model . This work stands out by its identification of a residue , tyrosine336 of huTRIM5α , which limits the spectrum of MLV targets of this antiviral to the sole N-tropic MLV . A number of amino acid substitutions at this position , as well as a deletion of this residue , confer huTRIM5α with the additional ability to block B-MLV , and introduction of a lysine even expands restriction to Mo-MLV . Understanding fully the mechanism of this gain of function would require a determination of the tri-dimensional structure of the TRIM5α-capsid complex . In its absence , the crystal structures of the PRYSPRY domain of related proteins , PRYSPRY-19q13 . 4 . 1 , GUSTAVUS and TRIM21 , suggests that the V1 region of huTRIM5α could form a protruding loop with tyrosine336 situated underneath and in direct contact ( at a less than 4 Å distance ) with residues in the V2 loop , which could limit the conformational flexibility of V1 [20–22] . However , the PRYSPRY V1 loops of TRIM5α and these other proteins differ in length , precluding firm analogy [23] . Alternatively , tyrosine336 may prevent the docking of V1 into its putative capsid-binding site by steric hindrance . This would be consistent with the finding that changes that most effectively broaden the spectrum of action of huTRIM5α are the removal of this tyrosine or its substitution by small amino acids . However , the observation that Y336K further expands the restriction spectrum of huTRIM5α not only to B- but also to Mo-MLV suggests that this model may be overly simplistic . Arginine332 of huTRIM5α was similarly found to interfere with ability of huTRIM5α to restrict HIV-1 and SIVmac [30] . In this case too , distinct amino acid changes differentially affected the strength with which either one of these two viruses was inhibited , suggesting that both positive and negative influences are at play . As well , our failure to generate a TRIM5α variant capable of blocking both N- and B-MLV on the one hand and HIV-1 on the other hand , by combining mutations at positions 332 and 336 , points to more complex influences within V1 itself . cis-acting interferences have also been noted in Fv1 , where the C-terminal part of Fv1b was shown to prevent this factor from blocking B-MLV , and where substitution of lysine358 of Fv1n by alanine could extend the restriction spectrum of this antiviral to N-MLV [36] . On the viral capsid side , our study indicates that at least four positions ( CA82 , CA109 , CA110 and CA117 ) interdependently condition MLV susceptibility to huTRIM5α , whether in its wild-type or mutant forms . The influence of each of these four residues varies according to both the virus involved and the sequence of the TRIM5α PRYSPRY V1 region . Here , all viruses tested escaped wild-type huTRIM5α if they harbored a glutamic acid at position 110 of capsid . As such , E110 dominantly interfere with restriction . However , a recent study demonstrated that wild-type huTRIM5α could efficiently block an MLV retroviral vector packaged with a capsid derived from a primate endogenous retrovirus ( PtERV ) carrying a glutamic acid at this position [37] . As well , we found here that the protective effect of E110 could be abrogated by substitutions of Y336 in huTRIM5α , in which case CA82 and CA117 became determinant . Indeed , with N- and B-MLV-derived viruses , an aspartate at CA82 additionally allowed escape from TRIM5αY336A , and a histidine at CA117 from TRIM5αY336A and TRIM5αY336K . When CA110 was occupied by an arginine , the picture was completely reversed , as this residue dominantly potentiates susceptibility . Finally , with an alanine at CA110 , H117 and D82 induced escape from wild-type TRIM5α , albeit in a CA109-dependent fashion , yet viruses remained sensitive to Y336-mutated forms of the restriction factor . A picture is thus emerging from these data , whereby CA110 plays the role of primary determinant of restriction , with CA82 , CA109 and CA117 acting as secondary modulators in a V1-conditioned fashion . However , the restriction pattern obtained with derivatives of Mo-MLV , which differs from the N- and B-MLV strains used here at nine CA positions besides these four , does not fully fit with this model , indicating its modulation by at least some of these other CA residues . Notably , a recent study demonstrated the importance of CA214 in potentiating Fv1n-mediated restriction of Mo-MLV only when CA110 was occupied by a glutamate [33] . MLV CA82 , CA109 , CA110 and CA117 were also demonstrated to exert combinatorial influences on Fv1-mediated restriction [35] . In spite of this parallel , sequences leading to resistance or susceptibility to Fv1 and huTRIM5α are not identical . For instance , whereas Fv1b and huTRIM5α both potently restrict CAE110A B-MLV , an additional N82D mutation allows escape from huTRIM5α ( this work ) but not from Fv1b [35] . Also , huTRIM5αY336A-mediated blockade of B-MLV is relieved by change at CA117 , which was previously shown not to affect restriction by Fv1n [35] . The structure of the amino-terminal part of the N-MLV capsid in its hexameric state was resolved at a 2 . 5 Å resolution [38] ( Figure 8 ) . A monomer consists of two-stranded β-hairpins followed by six α helices . Interestingly , CA82 , CA109 , CA110 and CA117 are situated at the edge of a cavity formed by helices 4 to 6 ( Figure 8 ) . CA82 sits between helix 4 and 5 at the top of this pocket , across from CA109 and 110 on helix 6 . CA117 is further down along the helix 6 side of the cavity . At least two scenarios can thus be envisioned for the binding of huTRIM5α to the MLV capsid . First , it might rely on the sum of individual interactions between TRIM5α residues , for instance in the PRYSPRY V1 loop , and capsid amino acids including 82 , 109 , 110 and 117 . The non-essential nature of any of these four capsid positions for susceptibility or resistance to TRIM5α argues against this model , even though it is conceivable that the abrogation of some of these interactions might be compensated by the strengthening of others . In a second scenario , the TRIM5α-binding site would be located deeper in the pocket . This part of the protein is constituted by residues that are highly conserved , hence most probably play essential structural functions prohibiting mutation [38] . Escape could then be achieved by mounting obstacles to TRIM5α penetration into this pocket through changes at the more flexible yet critically placed CA82 , CA109 , CA110 and CA117 residues . By analogy , it is interesting to note that the cyclophilin-binding loop of the HIV-1 capsid , which has been postulated to interfere with the blockade of this virus by wild-type huTRIM5α , hangs over a very similar pocket formed by helices 4 to 7 of the structurally homologous lentiviral capsid [38 , 39] . As such , this loop , whether bound to or modified by cyclophilin A ( CypA ) , could function as a lid to prevent huTRIM5α from accessing its HIV-1 CA binding site . However , recent data , which indicate that the positive effects of CypA binding to CA on HIV-1 replication do not depend upon the presence of huTRIM5α suggest that a strict parallel cannot be established between restriction of MLV and HIV by the cellular antiviral [40–42] .
Mus dunni tail fibroblasts ( MDTFs ) and human embryonic kidney 293T cells ( HEK 293T ) were purchased from the American Type Culture Collection ( ATCC ) . All cell lines were cultivated in Dulbecco's modified Eagle medium supplemented with 10% fetal calf serum , 2 mM glutamine and antibiotics ( 100 U/ml penicillin , 100 mg/ml streptomycin ) . MLV-based particles were produced using packaging constructs containing Moloney MLV ( pCIGPB ) , N- and B-tropic MLV CA ( pCIG3-N and pCIG3-B ) kindly provided by O . Danos and J . Stoye , respectively [32] . The GFP-encoding vector construct for all MLV reporter viruses was pCNCG kindly provided by R . Zufferey . Lentiviruses-based vectors were produced with the packaging construct psPAX2 ( Figures 2A and 3 ) or pR8 . 74 ( Figure 4 ) and the vector pWPTS-GFP ( Figures 2A and 3 ) or pRRLsin PGK GFP ( Figure 4 ) . The env construct for all viral productions was pMD2G plasmid expressing vesicular stomatitis virus G protein . Many plasmids used are distributed by Addgene ( http://www . addgene . org/ ) . The MLV plasmid encoding human TRIM5α was a kind gift of J . Sodroski and was already described [13] . The amino acid coding sequence of human TRIM5α with a C-terminal epitope derived from influenza virus hemagglutinin ( HA ) was inserted in pLPCX MLV vector construct ( Clontech ) allowing for puromycin selection of transduced cells . Site-directed mutagenesis on pLPCX-huTRIM5α-HA , pCIG3-N , pCIG3-B and pCIGPB was performed with the XL QuickChange mutagenesis kit from Stratagene . Primers used are listed in Table S1 . Proper site-directed mutagenesis was checked by sequencing reactions . All vector productions were performed by CaPO4-mediated transient co-transfection of the retroviral vector , gag-pol and env encoding constructs ( http://tronolab . epfl . ch/; with some minor adjustments ) . Briefly , subconfluent HEK 293T cells were co-transfected with 21 . 5 μg of vectors , 14 . 6 μg packaging constructs and 7 . 9 μg env constructs in a 15-cm plate . Cells were washed 16 hours post-transfection and supernatants were harvested 12 , 24 and 36 hours later . Recombinant retroviral vectors containing supernatants were centrifugated , filtrated , and in some cases were concentrated by ultracentrifugation . Titrations were performed on Fv1-null MDTF cells . MLV-based retroviral vectors encoding wild-type or point mutants of human TRIM5α were produced using the pLPCX-derived plasmids as described above . Viral supernatants containing recombinant retroviral vectors were added on 5 × 104 MDTF cells . Forty-eight hours post-transduction , cells were expanded and selection for stably transduced cells was performed by adding puromycin ( Sigma ) at a concentration of 5 μg/ml . Cells were maintained continuously in the presence of puromycin . To evaluate TRIM5α expression level , total proteins were extracted in a radioimmune precipitation assay buffer ( phosphate-buffered-saline ( PBS ) with 1% NP-40 , 0 . 5% sodium deoxycholate and 0 . 1% SDS ) supplemented with protease inhibitor cocktail ( Calbiochem ) . Equal amounts of protein were resolved on a Tris-glycine SDS-Polyacrylamide gel followed by western blot . HA-tagged proteins were detected using peroxydase-conjugated rat monoclonal antibody ( clone 3F10 , Roche ) . Proliferating cell nuclear antigen ( PCNA ) was used as a protein loading control and was detected using a mouse monoclonal antibody ( clone PC10 , Calbiochem ) followed by a secondary sheep anti-mouse antibody conjugated to horseradish peroxidase . MDTF cells stably expressing wild-type huTRIM5α , huTRIM5αY336A or stably transduced with the empty pLPCX construct as a control were seeded at 2 . 5 × 104 in a 24-well plate . N- , B- and Mo-MLV viral stocks encoding GFP were treated with DNAse I ( 20 μg/ml ) in the presence of MgCl2 ( 10 mM ) for 30 minutes at 37°C . Cells were then transduced at an equal low multiplicity of infection . For all time points and for each cell line , a PCR negative control with azidothymidine ( 62 . 5 μM , Calbiochem ) pre-treated cells was included . Cells were then harvested before transduction and 6 or 8 hours post-transduction . DNA was then extracted using the DNAeasy Tissue extraction kit from Quiagen . To detect the presence of provirus , cells were also collected 5 days post-transduction and processed for DNA extraction and FACS analysis . PCR reactions were performed using 5 μl of DNA extract . PCR amplified a region from the neomycin resistance gene ( forward primer: 5' GCGTTGGCTACCCGTGATATTG 3' ) to the cytomegalovirus promoter ( reverse primer: 5' TGGGCTATGAACTAATGACC 3' ) present in the intermediate reverse transcript resulting from RNA expressed by pCNCG . Mus musculus peripheral myelin protein ( Pmp22 , NM_008885 ) was used as a normalization gene ( forward primer: 5' TTCGTCAGTCCCACAGTTTTCTC 3' , reverse primer: 5' ACTCGCTAGTCCCAA GGGTCTA 3' ) . MDTF stable cell lines were seeded at 2 . 5 × 104 and transduced 24 hours later with 2-fold serial dilutions of GFP reporter vectors . Cells were harvested 48 hours post-transductions and fixed in 1% formaldehyde-containing PBS . The percentage of GFP-positive cells was determined by flow cytometry using the Beckton Dickinson FACScan or the multi-well plate reader Beckman Coulter Cell lab Quanta Flow Cytometer . Results were analysed with FlowJo 8 . 1 . 1 software . To calculate the fold restriction of the different MLV capsid mutants by huTRIM5α derivatives , a ratio was performed between the percentage of GFP-positive cells in the absence ( cells stably transduced with the empty vector ) and presence of huTRIM5α derivatives ( cells stably expressing wild-type or mutants huTRIM5α ) . Ratios were calculated with each dose of GFP vector from at least two independent infections , and the average of these ratios was used for the semi-quantitative scoring given in Figure 7 . The resolved structure of the N-terminal domain of N-MLV capsid in its hexameric state ( [38]; PDB: 1U7K ) was visualized using the UCSF Chimera software as described [43] .
The National Center for Biotechnology Information ( http://www . ncbi . nlm . nih . gov/ ) accession numbers for the proteins discussed in this paper are human TRIM5α ( AY625000 ) , peripheral myelin protein Pmp22 ( NM_008885 ) , and Rhesus TRIM5α ( AY523632 ) . | Mammalian cells are endowed with intrinsic lines of defence against retroviruses , which notably contribute to limiting the cross-species transmission of these pathogens . TRIM5α is one such restriction factor , which acts by recognizing the capsid of incoming retroviruses through its C-terminal PRYSPRY domain . Human TRIM5α potently blocks the so-called N-tropic murine leukemia virus ( MLV ) , but is ineffective against the closely related B-tropic and Moloney strains . In this study , we demonstrate that substitution of a single amino acid in the PRYSPRY domain of this protein expands its antiviral activity to these other MLV strains . Conversely , we show that protection of MLV from this restriction is governed by the negative influence of specific residues at a few critical positions of the retroviral capsid . These results support the model of a direct interaction between TRIM5α and retroviral capsids , shedding light on an important arm of innate antiretroviral immunity . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods",
"Supporting",
"Information"
] | [
"vertebrates",
"viruses",
"infectious",
"diseases",
"virology",
"eukaryotes",
"microbiology",
"mus",
"(mouse)",
"animals"
] | 2007 | Interfering Residues Narrow the Spectrum of MLV Restriction by Human TRIM5α |
Although they share certain biological properties with nucleic acid based infectious agents , prions , the causative agents of invariably fatal , transmissible neurodegenerative disorders such as bovine spongiform encephalopathy , sheep scrapie , and human Creutzfeldt Jakob disease , propagate by conformational templating of host encoded proteins . Once thought to be unique to these diseases , this mechanism is now recognized as a ubiquitous means of information transfer in biological systems , including other protein misfolding disorders such as those causing Alzheimer's and Parkinson's diseases . To address the poorly understood mechanism by which host prion protein ( PrP ) primary structures interact with distinct prion conformations to influence pathogenesis , we produced transgenic ( Tg ) mice expressing different sheep scrapie susceptibility alleles , varying only at a single amino acid at PrP residue 136 . Tg mice expressing ovine PrP with alanine ( A ) at ( OvPrP-A136 ) infected with SSBP/1 scrapie prions propagated a relatively stable ( S ) prion conformation , which accumulated as punctate aggregates in the brain , and produced prolonged incubation times . In contrast , Tg mice expressing OvPrP with valine ( V ) at 136 ( OvPrP-V136 ) infected with the same prions developed disease rapidly , and the converted prion was comprised of an unstable ( U ) , diffusely distributed conformer . Infected Tg mice co-expressing both alleles manifested properties consistent with the U conformer , suggesting a dominant effect resulting from exclusive conversion of OvPrP-V136 but not OvPrP-A136 . Surprisingly , however , studies with monoclonal antibody ( mAb ) PRC5 , which discriminates OvPrP-A136 from OvPrP-V136 , revealed substantial conversion of OvPrP-A136 . Moreover , the resulting OvPrP-A136 prion acquired the characteristics of the U conformer . These results , substantiated by in vitro analyses , indicated that co-expression of OvPrP-V136 altered the conversion potential of OvPrP-A136 from the S to the otherwise unfavorable U conformer . This epigenetic mechanism thus expands the range of selectable conformations that can be adopted by PrP , and therefore the variety of options for strain propagation .
Prion-mediated phenotypes and diseases result from the conformationally protean characteristics of particular amyloidogenic proteins . The prion state has the property of interacting with proteins in their non-prion conformation , thus inducing further prion conversion . The prion phenomenon has been described for a variety of different proteins involved in diverse biological processes ranging from translation termination in yeast , memory in Aplysia , antiviral innate immune responses [1] , and most recently the action of the p53 tumor suppressor [2] . Since the prion and non-prion conformations have differing biological properties , the net result of this replicative process is protein-mediated information transfer , the characteristics of which vary from prion to prion . The ubiquity of prion replication indicates that this is a wide-ranging of means of information transfer in biological systems . In the case of mammalian neurodegenerative diseases the prion state is pathogenic as well as transmissible . A hallmark of such conditions is the inexorable progression of pathology between synaptically connected regions of the central nervous system ( CNS ) , consistent with advancing cell-to-cell prion spread . Experimental transmission in several settings has been convincingly demonstrated in the case of the amyloid beta ( Aβ ) peptide which features prominently in Alzheimer's disease ( AD ) , the intracytoplasmic protein tau , also involved in AD as well as various neurodegenerative diseases referred to as taopathies , and α-synuclein , the primary constituent of Lewy bodies found in Parkinson's disease ( PD ) [1] , [3] . The prototypic and best-characterized prion diseases are the transmissible spongiform encephalopathies ( TSEs ) of animals and humans , including sheep scrapie , bovine spongiform encephalopathy ( BSE ) , chronic wasting disease ( CWD ) of cervids , and human Creutzfeldt-Jakob disease ( CJD ) . TSEs result from conformational conversion of the host-encoded cellular form of the prion protein , PrPC , to the corresponding prion , or scrapie form , PrPSc . Since TSEs share numerous properties with nucleic acid-based pathogens , including agent host-range , stable strain properties , and the ability to mutate and respond to selective pressure , early researchers assumed a viral etiology for these diseases . While this is not the case , the unequivocal infectivity of TSEs set these prions apart . Their singular capacity to cause fatal neurodegeneration in genetically tractable animal models , and the ability to propagate and quantify infectivity , in vivo , in cell culture or cell-free conditions , provide unparalleled settings to elucidate general mechanisms and devise integrated therapeutic approaches for all diseases involving conformational templating [4] . TSEs have long incubation periods ranging from months to years , are invariably fatal , and currently incurable . While a variant of CJD ( vCJD ) is unequivocally linked to prions causing BSE [5] , the zoonotic potential of other TSE's remains uncertain . Whereas all TSEs , including human genetic and sporadic forms , are experimentally transmissible , most are naturally infectious and frequently occur as unanticipated epidemics . Scrapie is one such example , and several iatrogenic epidemics have been reported . More than 1 , 500 sheep developed scrapie following administration of a scrapie-contaminated vaccine [6] . A similar recent event led to an ∼20-fold increase in the rate of scrapie in Italy [7] . Prion strain properties and the primary structure of PrP are the two major elements controlling prion transmission . Optimal disease progression appears to occur when the primary structures of PrPSc constituting the infectious prion , and substrate PrPC expressed in the host are closely related [8]–[10] . Underscoring the importance of primary structure on transmission , susceptibility and disease presentation are strongly influenced by several PRNP polymorphisms in humans and animals . For example , a strong association between susceptibility/resistance to natural scrapie is associated with the valine ( V ) /alanine ( A ) dimorphism at PrP residue 136 [11] . Prion strains are classically defined by differences in incubation times , and the neuropathological profiles they induce in the CNS . Seminal studies of mink prions [12] , as well as studies of human prions in Tg mice [13] indicated that strain information is enciphered within the tertiary structure of PrPSc . While this remains the favored explanation for prion strain diversity , the mechanism by which primary and higher order PrPC and PrPSc structures interact to influence pathogenesis are not understood . Our previous studies demonstrated that A at ovine PrP residue 136 is a component of the monoclonal antibody ( mAb ) PRC5 epitope [14] . This property allowed us to use PRC5 in this study to distinguish OvPrP-A136 from OvPrP-V136 , affording the opportunity to monitor allele-specific OvPrP conversion during prion infection . To accomplish this , we engineered Tg mice expressing either OvPrP-A136 or OvPrP-V136 , as well as Tg mice expressing both alleles in the same neuronal populations . Here , using a combination of in vivo and in vitro approaches , we address the mechanism by which this important disease susceptibility dimorphism influences scrapie strain-specific pathogenesis .
We created Tg mice expressing OvPrP encoding either A or V at residue 136 . Using semi-quantitative Western and immuno dot blotting we ascertained that levels of expression in the CNS of Tg ( OvPrP-A136 ) 3533+/− and Tg ( OvPrP-V136 ) 4166+/− mice were close to that of PrP expressed in the CNS of wild type mice ( Fig . 1A ) . Both lines of Tg mice tolerated these levels of expression without spontaneously developing recognizable signs of disease ( Table 1 ) . In contrast , Tg mice of both genotypes intracerebrally ( ic ) inoculated with brain homogenates from scrapie-affected sheep succumbed to the neurological effects of prion disease following variable incubation periods ( Table 1 ) . Rapid disease onset occurred following inoculation of Tg ( OvPrP-V136 ) 4166+/− mice with SSBP/1 prions [15] , [16] , which consistently produced an ∼130 d mean incubation time . While SSBP/1 also caused disease in Tg ( OvPrP-A136 ) 3533+/− mice , mean incubation times were ∼230 to 280 d longer ( Fig . 1B and Table 1 ) . In contrast , CH1641 prions [17] induced disease in all inoculated Tg ( OvPrP-A136 ) 3533+/− mice with a mean ∼310 d onset of disease , whereas no disease was registered in Tg ( OvPrP-V136 ) 4166+/− mice after >560 d . These distinct transmission profiles are consistent with previously recognized strain differences between SSBP/1 and CH1641 scrapie prions [17] . Consistent with this notion , western blot analysis of proteinase K-treated brain extracts of diseased Tg ( OvPrP-A136 ) 3533+/− mice confirmed that the molecular profiles which distinguish PrPSc constituting SSBP/1 and CH1641 prions [18] were maintained upon transmission ( Fig . 1C ) . These results demonstrate that Tg ( OvPrP-A136 ) 3533+/− and Tg ( OvPrP-V136 ) 4166+/− mice are capable of distinguishing scrapie strain-specific transmission patterns , and in turn that these properties are influenced by the A/V136 dimorphism . Previous studies revealed a positive correlation between PrPSc conformational stability and the incubation times of mouse and cervid prions [19] , [20] , but not of hamster prions [21] , [22] . We performed guanidine denaturation treatments on PrPSc in brain extracts of SSBP/1 infected Tg ( OvPrP-V136 ) 4166+/− mice with short incubation times and SSBP/1 infected Tg ( OvPrP-A136 ) 3533+/− mice with long incubation times . Analyses using mAb 6H4 revealed distinct stability curves for OvPrPSc-V136 and OvPrPSc-A136 . The conformational stability of OvPrPSc-V136 was lower than OvPrPSc-A136 in the range of GdnHCl concentrations between 1 and 2 M , ( Fig . 2A ) and GdnHCl1/2 values were 1 . 78 and 2 . 17 respectively . This confirmed that the conformation of OvPrPSc-V136 produced in Tg ( OvPrP-V136 ) 4166+/− mice with rapid incubation times was less stable than OvPrPSc-A136 produced in Tg ( OvPrP-A136 ) 3533+/− mice with longer incubation times . We refer to these conformations as unstable ( U ) and stable ( S ) , and to the rapidly and slowly propagating prions composed of these conformers as SSBP/1-V136 ( U ) , and SSBP/1-A136 ( S ) . We then used histoblotting [23] , a widely used method for characterizing strain-specific differences in PrPSc distribution [20] , [24] , with mAb 6H4 to characterize OvPrPSc-A136 ( S ) and OvPrPSc-V136 ( U ) deposition in the CNS . While OvPrPSc-A136 ( S ) had a punctate pattern of accumulation throughout the midbrain , pons , and oblongata of slow incubation time Tg ( OvPrP-A136 ) 3533+/− mice ( Fig . 3A ) , the neuroanatomical distribution of OvPrPSc-V136 ( U ) in the same sections of rapid incubation time Tg ( OvPrP-V136 ) 4166+/− mice was distinctly different , being more intense and diffusely deposited than OvPrPSc-A136 ( S ) ( Fig . 3B ) . Since Tg ( OvPrP-A136 ) 3533+/− and Tg ( OvPrP-V136 ) 4166+/− mice were both engineered using the same cosSHa . Tet cosmid vector which drives expression from the PrP gene promoter , we conclude that these differences are not the result of expression OvPrPC-A136 and OvPrPC-V136 in different neuronal populations . We used brain extracts of Tg ( OvPrP-A136 ) 3533+/− or Tg ( OvPrP-V136 ) 4166+/− mice as sources of OvPrPC-A136 and OvPrPC-V136 substrates for protein misfolding amplification ( PMCA ) [25] using SSBP/1 . While neither template spontaneously converted to PrPSc in the absence of seeded prions ( Fig . 4A and E ) , SSBP/1 reproducibly converted OvPrPC-V136 to OvPrPSc-V136 during a single round of PMCA ( Fig . 4C ) . In contrast , conversion was not observed after a single round of PMCA when OvPrPC-A136 was used as template ( Fig . 4B and Fig . 5A ) . We therefore used serial PMCA ( sPMCA ) [26] over 10 rounds to address whether conversion of OvPrPC-A136 to OvPrPSc-A136 might be detected after prolonged replication . Serial PMCA was performed in triplicate using equal amounts of PrP from three different Tg ( OvPrP-A136 ) 3533+/− or Tg ( OvPrP-V136 ) 4166+/− mouse brains using sheep SSBP/1 as seed . Apart from a slight but consistent decrease between rounds two and five , OvPrPSc-V136 production , detected using mAb 6H4 , was sustained throughout rounds one to 10 ( Fig . 4C ) . As expected , mAb PRC5 failed to detect OvPrPSc-V136 ( Fig . 4G ) . In contrast , OvPrPSc-A136 was undetectable with mAbs 6H4 or PRC5 until round eight , after which levels decreased during rounds nine and 10 ( Fig . 4B and F ) . Having established that Tg mice expressing OvPrPC-V136 and OvPrPC-A136 propagate SSBP/1-V136 ( U ) and SSBP/1-A136 ( S ) prions with relatively rapid and slow incubation times respectively , we produced Tg ( OvPrP-A/V136 ) mice expressing both OvPrPC-A136 and OvPrPC-V136 and inoculated them with SSBP/1 to examine whether disease developed with fast , slow or intermediate kinetics . Although more rapid than the ∼130 d onset of disease in Tg ( OvPrP-V136 ) 4166+/− mice ( P = 0 . 0094 ) , the mean 105±5 d onset of disease contrasted with the >400 d SSBP/1 incubation times observed in Tg ( OvPrP-A136 ) 3533+/− mice ( Fig . 1B and Table 1 ) . Stability assessments using mAb 6H4 showed that the denaturation curves of OvPrPSc produced in the brains of diseased Tg ( OvPrP-V136 ) 4166+/− and Tg ( OvPrP-A/V136 ) mice were superimposable over most of the range of GdnHCl concentrations ( Fig . 2A ) , indicating that OvPrPSc produced in Tg ( OvPrP-A/V ) mice shared the conformation of OvPrPSc-V136 ( U ) produced in SSBP/1 infected Tg ( OvPrP-V136 ) 4166+/− mice . In accordance with this notion , histoblotting using mAb 6H4 showed that the neuroanatomical distribution of OvPrPSc ( U ) in the brains of diseased Tg ( OvPrP-A/V ) mice mirrored the diffuse deposition of the OvPrPSc-V136 ( U ) conformer located in similar sections of rapid incubation time Tg ( OvPrP-V136 ) 4166+/− mice ( Fig . 3C ) . While the rapid SSBP/1 incubation times , and properties of the converted PrPSc in diseased Tg ( OvPrP-A/V ) were consistent with propagation of SSBP/1-V136 ( U ) prions , remarkably , western blotting of diseased Tg ( OvPrP-A/V136 ) brain extracts with mAb PRC5 revealed substantial conversion of OvPrC-A136 to OvPrPSc-A136 ( Fig . 2D ) . Densitometric comparisons of OvPrPSc levels using mAbs 6H4 ( Fig . 2B ) and PRC5 ( Fig . 2D ) allowed us to estimate relative conversion efficiencies of each allele product in the brains of SSBP/1 infected Tg ( OvPrP-A/V136 ) mice . Using samples from diseased Tg ( OvPrP-A136 ) 3533+/− mice probed with mAbs 6H4 and PRC5 as normalizing controls for differences in the affinities of the two mAbs for OvPrP-A136 , we estimated by Western or dot blotting that OvPrPSc-A136 comprised ∼45% of total PK-resistant PrP in the brains of diseased Tg ( OvPrP-A/V136 ) mice . We then used mAb PRC5 to determine the conformation of OvPrPSc-A136 among total OvPrPSc produced in the brains of diseased Tg ( OvPrP-A/V136 ) mice . The 1 . 64 GdnHCl1/2 value of OvPrPSc-A136 produced under these conditions was distinct from that of OvPrPSc-A136 ( S ) produced in long incubation time Tg ( OvPrP-A136 ) 3533+/− mice ( GdnHCl1/2 = 2 . 11 ) , and their non-superimposable PRC5 denaturation curves were significantly different in the range of 1 . 5–2 . 5 M GdnHCl ( Fig . 2C ) . These findings demonstrated that the conformation of OvPrPSc-A136 in rapid incubation time Tg ( OvPrP-A/V ) mice was distinct from OvPrPSc-A136 ( S ) produced in long incubation time Tg ( OvPrP-A136 ) 3533+/− mice . We refer to this novel conformation as OvPrPSc-A136 ( U ) , and to the resulting prions as SSBP/1-A136 ( U ) . Histoblotting using mAb PRC5 confirmed the comparatively limited and punctate distribution pattern of OvPrPSc-A136 ( S ) in the CNS of long incubation time Tg ( OvPrP-A136 ) 3533+/− mice ( Fig . 3D ) that we observed with mAb 6H4 ( Fig . 3A ) . As expected , OvPrPSc-V136 ( U ) in the CNS of diseased Tg ( OvPrP-V136 ) 4166+/− mice was refractory to detection by mAb PRC5 ( Fig . 3E ) . We probed histoblots of the CNS from diseased Tg ( OvPrP-A/V136 ) mice with mAb PRC5 to assess the appearance and distribution of OvPrPSc-A136 ( U ) . In contrast to the punctate deposits of OvPrPSc-A136 ( S ) in long incubation time Tg ( OvPrP-A136 ) 3533+/− mice ( Fig . 3A and D ) , OvPrPSc-A136 ( U ) in Tg ( OvPrP-A/V136 ) mice ( Fig . 3F ) acquired a diffuse deposition and a distribution pattern that was equivalent to OvPrPSc-V136 ( U ) in Tg ( OvPrP-V136 ) 4166+/− mice ( Fig . 3B ) . Consistent with the co-expression of each allele in identical cell populations of Tg ( OvPrP-A/V ) mice , spatial distributions of 6H4- and PRC5-reactive PrP coincided in all analyzed sections of Tg ( OvPrP-A/V ) mice ( Fig . 3G and H ) . To simulate the combined effects of OvPrP-A136 and OvPrP-V136 on PrP conversion in Tg ( OvPrP-A/V136 ) mice in vitro , we mixed equal quantities of OvPrPC-A136 and OvPrPC-V136 in PMCA reactions seeded with SSBP/1 . Under these conditions , similar to when OvPrPC-V136 was present in isolation ( Fig . 4C ) , we observed early , reproducible conversion to OvPrPSc in round one ( Fig . 4D ) . Probing of western blots with mAb PRC5 showed that OvPrPSc-A136 was a component of this converted material ( Fig . 4H ) . Thus , similar to our observations in Tg mice , the presence of OvPrPC-V136 induced the relatively rapid conversion of OvPrPC-A136 to OvPrPSc-A136 by SSBP/1 . Interestingly , subsequent conversion of both OvPrPC-A136 and OvPrPC-V136 diminished in rounds two to five , ultimately becoming undetectable through rounds six to 10 ( Figs . 4D and H ) . SSBP/1 was originally produced from a pool of diseased sheep brains from the positive selection line in the Neuropathogenesis Unit ( NPU ) Cheviot sheep flock , and has subsequently been passaged as a pool . We next compared the seeding properties of SSBP/1 with those of SSBP/1-A136 ( S ) or SSBP/1-V136 ( U ) prions derived from SSBP/1-infected Tg ( OvPrP-A136 ) 3533+/− or Tg ( OvPrP-V136 ) 4166+/− mice . We monitored conversion of OvPrPC-A136 or OvPrPC-V136 templates every two hours for a total of 12 h of PMCA . SSBP/1-V136 ( U ) had the same PMCA properties as SSBP/1: both SSBP/1 and SSBP/1-V136 ( U ) prions efficiently converted OvPrPC-V136 in isolation , but not OvPrPC-A136 in isolation; when both templates were present in the PMCA reaction , the presence of OvPrPC-V136 facilitated conversion of OvPrPC-A136 to OvPrPSc-A136 by SSBP/1 or SSBP/1-V136 ( U ) prions ( Figs . 5A and B ) . In contrast , SSBP/1-A136 ( S ) prions converted either OvPrPC-V136 or OvPrPC-A136 templates to PrPSc when they were present in isolation , the latter being unequivocally confirmed to be OvPrPSc-A136 using mAb PRC5 ( Fig . 5C ) ; however , in the presence of both templates SSBP/1-A136 ( S ) prion propagation was inhibited ( Fig . 5C ) . The properties of prions derived from Tg ( OvPrP-A/V ) mice differed from SSBP/1 , SSBP/1-A136 ( S ) or SSBP1/-V136 ( U ) prions . Like SSBP/1 and SSBP/1-V136 ( U ) , prions passaged through these mice efficiently converted OvPrPC-V136 , but not OvPrPC-A136 . However , unlike SSBP/1 and SSBP/1-V136 ( U ) , such prions failed to facilitate conversion of OvPrPC-A136 in the presence of OvPrPC-V136 ( Fig . 5D ) .
Previous studies described the production of Tg mice expressing OvPrP , and reported their susceptibility to scrapie prions [27]–[32] . The most widely characterized models are tg338 mice expressing OvPrP-V136 [31] , and Tgov59 [33] or Tgov4 [29] lines expressing OvPrP-A136 . In the case of tg338 mice , the transgene was comprised of a bacterial artificial chromosome insert of 125 kb of sheep DNA , while in the case of Tgov59 and Tgov4 mice the neuron specific enoloase promoter was used to drive OvPrP expression . These lines are maintained on different heterogeneous genetic backgrounds , and CNS expression levels in tg338 mice are ∼8- to 10-fold higher than wild type , while Tgov59 and Tgov4 lines each over express OvPrP-ARQ at levels ∼2- to 4-fold higher than those found in sheep brain . Spontaneous neurological dysfunction has been reported in Tg lines over expressing OvPrP [27] , [31] . Tg ( OvPrP-A136 ) 3533+/− and Tg ( OvPrP-V136 ) 4166+/− mice express transgene-encoded PrP , either slightly lower , or slightly higher than PrP levels normally expressed in the CNS of wild type mice . Since Tg ( OvPrP-A136 ) 3533+/− and Tg ( OvPrP-V136 ) 4166+/− lines were produced using the cosSHa . Tet cosmid vector which drives expression from the PrP gene promoter [34] , we expected expression of OvPrPC-A136 and OvPrPC-V136 in identical neuronal populations , and therefore that both alleles are co-expressed in the same cells of Tg ( OvPrP-A/V ) mice . Finally , other than variable transgene insertion loci , both lines are otherwise sygeneic on an inbred Prnp0/0/FVB background . Previous studies reported on Tg mice expressing OvPrP with V at 136 , referred to as Tg ( OvPrP ) 14882+/− mice , that were also produced in a Prnp0/0/FVB background using the cosSHa . Tet cosmid vector [32] . However , in that study , comparable Tg mice expressing OvPrP-A136 were not reported . Median SSBP/1 scrapie incubation times in Tg ( OvPrP ) 14882+/−mice were 75 d , and this line expresses OvPrP at levels only slightly higher than Tg ( OvPrP-V136 ) 4166+/− mice . While we exercise caution when comparing results from mice produced by different groups , the otherwise similar properties of Tg ( OvPrP ) 14882+/− and Tg ( OvPrP-V136 ) 4166+/− mice suggest that even slight differences in the levels of transgene expression can have significant effects on prion incubation time . A clear link to codon 136 genotype and susceptibility/resistance to different sheep scrapie isolates has been described in multiple previous studies . Importantly , the influence of residue 136 on the transmission of SSBP/1 and CH1641 prions in Tg ( OvPrP-A136 ) 3533+/− and Tg ( OvPrP-V136 ) 4166+/− mice is in accordance with the properties of these isolates in sheep of various genotypes [17] . Generally , increased susceptibility to scrapie is associated with expression of OvPrP-V136 , with A/A136 being the most resistant , and V/V136 the most susceptible genotypes . In the case of SSBP/1 incubation periods are ∼170 days in V/V136 sheep , while transmission to A/A136 sheep is relatively inefficient , with no disease recorded after >1000 days [35] . While SSBP/1 eventually transmits to Tg ( OvPrP-A136 ) 3533+/− mice with incubation times exceeding 400 days , the general effects of the A/V136 dimorphism on SSBP/1 transmission observed in sheep are recapitulated in Tg ( OvPrP-A136 ) 3533+/− and Tg ( OvPrP-V136 ) 4166+/− mice ( Table 1 ) . Similarly , CH1641 , which propagates efficiently in A/A136 sheep [35] , preferentially propagates in Tg ( OvPrP-A136 ) 3533+/− mice ( Table 1 ) . In previous studies , CH1641 transmitted to TgOvPrP4 mice with an ∼250 d mean incubation time [36] . Although SSBP/1 incubation times are prolonged in A/V136 compared to V/V 136 sheep [35] , in our studies incubation times were shorter in Tg ( OvPrP-A/V ) than in Tg ( OvPrP-V136 ) 4166+/− mice . While the condition of A/V136 heterozygosity has not been previously modeled in Tg mice , this difference may result from double the levels of transgene expression in Tg ( OvPrP-A/V ) mice compared to Tg ( OvPrP-A136 ) 3533+/− and Tg ( OvPrP-V136 ) 4166+/− mice . Tg ( OvPrP-A/V136 ) mice were derived by mating Tg ( OvPrP-A136 ) 3533+/+ with Tg ( OvPrP-V136 ) 4166+/+ mice , and therefore express greater total levels of OvPrP than Tg ( OvPrP-A136 ) 3533+/− and Tg ( OvPrP-V136 ) 4166+/− mice ( Fig . 1A ) . Since the levels of OvPrP-V136 are equivalent in Tg ( OvPrP-V136 ) 4166+/− and Tg ( OvPrP-A/V ) mice , and we show that OvPrP-A136 also becomes available for conversion , this situation results in more available substrate for conversion . While previous studies revealed an inverse correlation between transgene expression levels and prion incubation times in Tg mice [8] , whether shorter incubation periods in Tg ( OvPrP-A/V136 ) mice than in Tg ( OvPrP-V136 ) 4166+/− mice reflect overall differences in PrPC expression levels remains uncertain . Differences in scrapie pathogenesis between mice and sheep may also reflect the influence of additional factors on disease in the natural host including other PRNP polymorphisms [37] , [38] , and different involvements of the lymphoreticular system in sheep compared to Tg mice . Our observations in Tg mice expressing individual allele products suggested that rapid or prolonged SSBP/1 incubation times in Tg ( OvPrP-V136 ) 4166+/− and Tg ( OvPrP-A136 ) 3533+/− mice respectively , reflected preferential conversion by SSBP/1 prions of OvPrPC-V136 , rapidly producing a relatively unstable OvPrPSc-V136 ( U ) conformation that was diffusely deposited in the CNS , compared to the slower conversion of OvPrPC-A136 to the more stable OvPrPSc-A136 ( S ) conformer which accumulated in the CNS with a punctate pattern ( Figs . 1–3 ) . Our results are consistent with the selection by the A/V136 dimorphism of SSBP/1-A136 ( S ) and SSBP/1-V136 ( U ) prions in Tg ( OvPrP-A136 ) 3533+/− and Tg ( OvPrP-V136 ) 4166+/− mice respectively . We also show that PMCA recapitulates the influence of the A/V136 polymorphism on the kinetics of SSBP/1 propagation observed in Tg mice . The general conclusions from these studies agree with previously published assessments of the mechanism of conformational selection by distinct PrP primary structures [39] , [40] . Based on the rapid SSBP/1 incubation times in Tg ( OvPrP-A/V136 ) mice , and shared conformational and distribution properties of OvPrPSc produced under these conditions with OvPrPSc-V136 ( U ) in Tg ( OvPrP-V136 ) 4166+/− mice , we speculated that OvPrP-A136 played no part during the propagation of SSBP/1 prions in Tg ( OvPrP-A/V ) mice . To address this we used mAb PRC5 to exclusively monitor conversion of OvPrPC-A136 . Surprisingly , in contrast to its relatively slow conversion when OvPrPC-A136 is expressed in isolation , co-expression with OvPrPC-V136 in Tg ( OvPrP-A/V136 ) mice facilitated rapid conversion of OvPrPC-A136 to OvPrPSc-A136 . The conformation and diffuse CNS distribution of the resulting OvPrPSc-A136 ( U ) were equivalent to that of OvPrPSc-V136 ( U ) and not OvPrPSc-A136 ( S ) . Collectively , these results lead us to conclude that once OvPrPSc-V136 ( U ) is formed by conversion of OvPrPC-V136 by SSBP/1 prions , the resulting unstable conformation induces rapid conversion of OvPrPC-A136 to OvPrPSc-A136 ( U ) . That this outcome is dependent on allele co-expression within the host is demonstrated by the inability of the OvPrPSc-V136 ( U ) conformer to template OvPrPC-A136 when it is expressed in isolation . Effects of OvPrP genotype on the propagation of scrapie prions were not controlled during the isolation and propagation of SSBP/1 . Passage of SSBP/1 through Tg mice therefore allowed us to generate prions composed solely of OvPrPSc-A136 , OvPrPSc-V136 , or mixtures of both , and to draw additional conclusions about the effects of the A/V136 dimorphism of prion propagation using PMCA . Similar to our observations in Tg mice , SSBP/1 failed to convert OvPrPC-A136 to OvPrPSc-A136 by PMCA , except in the presence of OvPrPC-V136 ( Figs . 4 and 5 ) . These conversion properties are shared with SSBP/1-V136 ( U ) prions , but are distinct from SSBP/1-A136 ( S ) prions , which show facile conversion of both OvPrPC-V136 and OvPrPC-A136 . These results suggest that the SSBP/1-V136 ( U ) is the dominant strain in the natural SSBP/1 isolate . Multiple parameters could account for this , including , but not restricted to , the effects of OvPrP genotype , for example as a result of exclusive propagation in sheep of the V136/V136 genotype , route of transmission in the infected sheep , and differential/selective prion replication in the lymphoreticular or central nervous systems of sheep . While our analyses indicate the presence of both OvPrP-V136 and OvPrP-A136 alleles in SSBB/1 ( Table 1 ) , it is important to note that SSBP/1 was derived from a pool of sheep brains of undefined genotypes . PCR approach precludes assessment of the extent to which alleles are present in a sample , raising the possibility that the one or other allele exists as a minor component in SSBP/1 . Our findings also suggest that PrPSc conformers may cross-inhibit PrP conversion . In case of SSBP/1-A136 ( S ) prions , the presence of OvPrPC-V136 inhibited PMCA of OvPrPC-A136 ( Fig . 5C ) . Also , while SSBP/1 seeding of PMCA reactions containing mixtures of OvPrPC-A136 and OvPrPC-V136 resulted in robust , reproducible conversion to OvPrPSc-A136 as early as round one ( Fig . 4H ) , total PrPSc production was ephemeral with subsequent PrPSc formation diminishing during rounds two to five , and conversion ultimately becoming undetectable after round six . Since early PrPSc conversion was sustained out to round 10 when OvPrPSc-A136 was not produced ( Fig . 4C and G ) , these results are consistent with inhibited conversion of OvPrPC-V136 to OvPrPSc-V136 by OvPrPSc-A136 . While early ( round one ) PMCA conversion of PrPSc by SSBP/1 with either OvPrPC-V136 or mixtures of OvPrPC-V136 and OvPrPC-A136 correlates with early onset of disease following SSBP/1 infection of both Tg ( OvPrP-V136 ) and Tg ( OvPrP-A/V136 ) mice , the subsequent inhibitory effects of OvPrPSc-A136 observed in PMCA would be impossible to detect in vivo , since Tg ( OvPrP-A/V ) mice succumb to the lethal effects of early PrPSc accumulation . Consistent with an inhibitory effect of OvPrPSc-A136 ( U ) , prions from Tg ( OvPrP-A/V ) mice , while they converted OvPrPC-V136 in isolation , failed to convert OvPrPC-A136 to PrPSc in the presence of OvPrPC-V136 ( Fig . 5D ) . Thus , the properties of prions from this defined genetic background differ from SSBP/1 . We emphasize that , despite PCR data supporting the presence of OvPrP-A136 alleles in this isolate , SSBP/1 was derived from sheep of undefined OvPrP genotypes , rather than sheep with a defined heterozygous OvPrP-A/V136 genotype . The inter-related effects of PrP primary and higher order structures on prion transmission were addressed in the Conformational Selection Model , which proposed that strains are composed of a range of PrPSc conformers , or quasi-species , and that only a subset of PrPSc conformations is compatible with each PrP primary structure [41] . While this model also took into account the effects of polymorphic variation on prion propagation , it did so only in the context of Tg mice expressing individual PrP allele products . Transgenetic studies of the human codon 129 methionine ( M ) /valine ( V ) polymorphism , and the analogous codon 132 M/leucine ( L ) polymorphism in elk , indicated that these dimorphisms acted to restrict or promote the propagation of particular prion strains [39] , [40] . While the responses of Tg ( OvPrP-A136 ) 3533+/− and Tg ( OvPrP-V136 ) 4166+/− mice are consistent with this notion , that is selection of the U conformer by OvPrP-V136 , and the S conformer by OvPrP-A136 , our unprecedented ability to analyze allele specific conversion in infected Tg ( OvPrP-A/A136 ) mice reveals a more complex mechanism where mixtures of PrP variants may assist or inhibit the propagation of strains under various conditions . For example , SSBP/1 or SSBP/1-V136 ( U ) prions facilitate conversion of OvPrPC-A136 to OvPrPSc-A136 ( U ) only in the presence of OvPrPC-V136 . Expressed in isolation , conversion of OvPrPC-A136 is favored by the OvPrPSc ( S ) conformer . Our results demonstrate that co-expression of different polymorphic forms of PrP , which would be the norm in humans and animals , have profound effects on conformational selection of prion strains . The results reported here address the molecular mechanisms associated with the phenomenon of prion strain over-dominance first observed by Dickinson and Outram [42] , and subsequently reported in other settings involving co-expression of long and short incubation time PrP alleles [43] . While this phenomenon was reconciled at the time by the assumption that TSEs were caused by unidentified viral agents , our results now indicate that the suggestion raised by those studies , namely that over-dominance most likely resulted from physical interaction of allele products of the scrapie incubation time locus during infection , was prescient . Our results support a molecular mechanism involving cross templating of an otherwise resistant allele product by a dominant prion conformer , in this case OvPrPSc ( U ) , which , we speculate , involves physical association of otherwise “susceptible” and “resistant” allele products . Consistent with the observations reported here , prion strain interference may also utilize similar mechanisms of conformational selection in a host expressing different PrP allele products infected with long and short incubation period strains with different PrPSc conformational stabilities [44] , [45] . In conclusion , we have used a combination of transgenic , immunologic , and in vitro approaches to explore the mechanism by which PrP primary structure variations and the conformations enciphered by different prion strains interact to control TSE propagation . While our results support previous studies indicating that PrP susceptibility polymorphisms , expressed in isolation , act to restrict or promote the propagation of particular prion conformers , we now show that under conditions of allele co-expression a dominant conformer may alter the conversion potential of an otherwise resistant PrP polymorphic variant to an unfavorable prion strain . While such responses are analogous to the phenotypic expression of genetically determined heritable traits , dominant prion conformers act epigenetically by means of protein-mediated conformational templating . By expanding the range of possible conformations adoptable by a particular prion protein primary structure , such interactive effects provide a mechanism for promoting strain fitness , and , we speculate , strain diversification . While the precise number scrapie strains in sheep and goats remains uncertain , the description of at least 24 additional major sheep PRNP polymorphisms , and combinations thereof , is likely to have a significant influence on strain diversity .
All animal work was conducted according to the National Institutes of Health guidelines for housing and care of laboratory animals , and performed under protocols approved by the Colorado State University Institutional Animal Care and Use Committee , with approval number 11-2996A . Sequences upstream of codon 44 of the OvPrP-A136 and V136 coding sequences were replaced with the corresponding sequence from mouse PrP . The resulting constructs contained the OvPrP coding sequence , except for addition of an extra residue for glycine at codon 31 , and the mouse PrP N-terminal signal peptide instead of OvPrP signal peptide . Tg mice were generated by cloning the OvPrP-A136 and OvPrP-V136 expression constructs into the cosSHa . Tet cosmid vector [34] , and microinjection of embryos from inbred Prnp0/0/FVB mice . Tg founders were identified by PCR screening of genomic DNA isolated from tail snips . Founder mice were mated with inbred Prnp0/0/FVB mice , and generally maintained with the transgene in the hemizygous state , with Tg mice identified by PCR screening of genomic DNA from weanlings . It was also possible to generate homozygous counterparts of each line , and Tg ( OvPrP-A/V136 ) mice were generated by crossing homozygous Tg ( OvPrP-V136 ) 4166+/+ mice with homozygous Tg ( OvPrP-A136 ) 3533+/+ mice . We used immuno-dot blotting and Western blotting with mAb 6H4 ( Prionics , Schlieren , Switzerland ) to estimate the levels of OvPrP expression . Tg mice subsequently shipped to and maintained in Edinburgh were crossed onto the Prnp0/0/129Ola background [46] . SSBP/1 originated as a homogenate of three natural scrapie brains that were subsequently passaged mostly through Cheviot sheep at the Neuropathogenesis Unit ( NPU ) , Edinburgh UK [15] , [16] . CH1641 is a naturally infected cheviot sheep from the NPU flock [17] . The presence of OvPrP-A136 or OvPrP-V136 alleles in these samples was ascertained by restriction fragment length polymorphism analysis of the PCR amplified PRNP coding sequences . Ten % mouse brain homogenates ( w/v ) were prepared in phosphate-buffered saline ( PBS ) lacking calcium and magnesium ions by repeated extrusion through 18- and 21-gauge needles . Sheep brain homogenates ( 10% ) in PBS were prepared by repeated extrusion through 14-gauge , followed by 18- to 28-gauge needles in PBS . Total protein content was determined by bicinchonic acid ( BCA ) assay ( Pierce Biotechnology , Inc . ) . Anesthetized mice were inoculated intracerebrally with 30 µl of 1% ( w/v ) brain extracts prepared and diluted in PBS . General health was monitored daily . Onset of prion disease was determined by observation of the progressive development of at least three of the following clinical signs: truncal ataxia , loss of extensor reflex , difficulty righting from a supine position , plastic tail , head bobbing or tilting , kyphotic posture , circling and paresis/paralysis . Animals were diagnosed when at least two investigators agreed with the manifestation of these signs . Incubation time is defined as the period between the time of inoculation to the day on which subsequently progressive clinical signs were initially recorded . Brain homogenates containing 500 µg protein were digested with 400 µg/ml proteinase K ( PK ) in 0 . 4 M NaCl , 10 mM Tris–HCl , pH 8 . 0 , 2 mM EDTA , pH 8 . 0 , and 2% SDS at 55°C overnight . Genomic DNA was precipitated with isopropanol . The partial OvPrP coding sequence was amplified by PCR with the forward and reverse primers: 5′-GGACAGGGCAGTCCTGGA-3′ , 5′-GTGATGCACATTTGCTCCACCACT-3′ . PCR products were purified with QIAquick Gel Extraction kit ( QIAGEN Science , MA , USA ) , digested with BspH I that only recognizes the OvPrP-V136 allele , and the products were resolved on a 1 . 2% agarose gel . Tg mice were perfused with PBS/5 mM EDTA . Ten % brain homogenates ( w/v ) were prepared in PBS containing 150 mM NaCl , 1 . 0% Triton X-100 , and the complete TM cocktail of protease inhibitors ( Roche , Mannheim , Germany ) . Samples were clarified by brief , low-speed centrifugation . Protein concentrations of brain homogenates used as substrates for PMCA were adjusted to contain equivalent amounts of OvPrP-A136 or OvPrP-V136 , based on the estimated relative levels of transgene expression . Substrates in which OvPrP-A136 and OvPrP-V136 were mixed were adjusted based on the estimated relative levels of transgene expression , so that approximately equal amounts of each allele product were present in the PMCA reaction . PMCA reactions were performed as described previously [20] , [47] at a seed to substrate ratio of 1∶180 . One cycle corresponded to 20 seconds of sonication followed by 30 minutes incubation at 37°C . Controls samples were incubated for the same duration at 37°C without sonication . Amplified and control samples were digested with PK at a final concentration of 0 . 33 µg/µl and analyzed on western blots using mAbs 6H4 or PRC5 . Brain homogenates and cell lysates were digested with 100 µg/ml or 30 µg/ml of PK respectively ( Roche , Mannheim , Germany ) in cold lysis buffer for 1 h at 37°C . Digestion was terminated with phenylmethylsulfonyl fluoride at a final concentration of 2 µM . Samples were boiled for 10 min in the absence of β-meracaptoethanol [14] and proteins were resolved by SDS-PAGE and transferred to polyvinylidenedifluoride Immobilon ( PVDF ) -FL membranes ( Millipore , Billerica , USA ) . Membranes were probed with primary mAbs followed by horseradish peroxidase–conjugated anti-mouse secondary antibody ( GE Healthcare , Little Chalfont , UK ) . Protein was visualized by chemiluminescence using ECL Plus ( GE Healthcare , Piscataway , USA ) and an FLA-5000 scanner ( Fujifilm Life Science , Woodbridge , USA ) . Brain homogenates containing 5 µg protein were incubated with various concentrations of guanidine hydrochloride ( GdnHCl ) in 96-well plates for 1 h at room temperature . Samples were adjusted with PBS to a final of concentration of GdnHCl of 0 . 5 M and transferred onto nitrocellulose ( Whatman GmbH , Dassel , Germany ) using a dot blot apparatus . After two PBS washes , the membrane was air-dried for 1 h , then incubated with 5 µg/mL PK in 50 mM Tris-HCl , pH 8 . 0 , 150 mM NaCl , 0 . 5% sodium deoxycholate , 0 . 5% Igepal CA-630 for 1 h at 37°C . PK was inactivated with 2 mM PMSF . The membrane was incubated in 3 M guanidine thiocyanate in Tris-HCl , pH 7 . 8 for 10 min at room temperature . After four washes with PBS , the membrane was blocked with 5% nonfat milk in TBST for 1 h , and probed with mAbs 6H4 ( 1∶20 , 000 ) or PRC5 ( 1∶5000 ) overnight at 4°C , followed by HRP-conjugated goat anti-mouse IgG secondary antibody . The membrane was developed with ECL Plus and scanned with GE image quant 4000 . The signal was analyzed with ImageQuant TL 7 . 0 software . Histoblots were produced and analyzed according to previously described protocols [23] . Images were captured with a NikonDMX 1200F digital camera in conjunction with Metamorph software ( Molecular Devices ) . | Prions are infectious proteins , originally discovered as the cause of a group of transmissible , fatal mammalian neurodegenerative diseases . Propagation results from conversion of the host-encoded cellular form of the prion protein to a self-propagating disease-associated conformation . It is believed that the self-propagating pathogenic form exists in a variety of subtly different conformations that encipher prion strain information . Here we explored the mechanism by which prion protein primary structural variants , differing at only a single amino acid residue , interact with prion strain conformations to control disease phenotype . We show that under conditions of co-expression , a susceptible prion protein variant influences the ability of an otherwise resistant variant to propagate an otherwise unfavorable prion strain . While this phenomenon is analogous to the expression of genetically-determined phenotypes , our results support a mechanism whereby dominant and recessive prion traits are epigenetically controlled by means of protein-mediated conformational templating . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [] | 2013 | Epigenetic Dominance of Prion Conformers |
Directed cell motion in response to an external chemical gradient occurs in many biological phenomena such as wound healing , angiogenesis , and cancer metastasis . Chemotaxis is often characterized by the accuracy , persistence , and speed of cell motion , but whether any of these quantities is physically constrained by the others is poorly understood . Using a combination of theory , simulations , and 3D chemotaxis assays on single metastatic breast cancer cells , we investigate the links among these different aspects of chemotactic performance . In particular , we observe in both experiments and simulations that the chemotactic accuracy , but not the persistence or speed , increases with the gradient strength . We use a random walk model to explain this result and to propose that cells’ chemotactic accuracy and persistence are mutually constrained . Our results suggest that key aspects of chemotactic performance are inherently limited regardless of how favorable the environmental conditions are .
Chemotaxis plays a crucial role in many biological phenomena such as organism development , immune system targeting , and cancer progression [1–4] . Specifically , recent studies indicate that chemotaxis occurs during metastasis in many different types of cancer [2 , 5–9] . At the onset of metastasis , tumor cells invade the surrounding extracellular environment , and oftentimes chemical signals in the environment can direct the migration of invading tumor cells . Several recent experiments have quantified chemotaxis of tumor cells in the presence of different chemoattractants [3] and others have been devoted to the intracellular biochemical processes involved in cell motion [10] . Since the largest cause of death in cancer patients is due to the metastasis , it is important to understand and prevent the directed and chemotactic behavior of invading tumor cells . Chemotaxis requires sensing , polarization , and motility [11] . A cell’s ability to execute these interrelated aspects of chemotaxis determines its performance . High chemotactic performance can be defined in terms of several properties . Cell motion should be accurate: cells should move in the actual gradient direction , not a different direction . Cell motion should be persistent: cells should not waste effort moving in random directions before ultimately drifting in the correct direction . Cell motion should be fast: cells should arrive at their destination in a timely manner . Indeed , most studies of chemotaxis use one or more of these measures to quantify chemotactic performance . Accuracy is usually quantified by the so-called chemotactic index ( CI ) , most often defined in terms of the angle made with the gradient direction [12–15] ( Fig 1A ) ; although occasionally it is defined in terms of the ratio of distances traveled [16] or number of motile cells [17–19] in the presence vs . absence of the gradient . Directional persistence [10] ( DP ) is usually quantified by the ratio of the magnitude of the cell’s displacement ( in any direction ) to the total distance traveled by the cell ( Fig 1A; sometimes called the McCutcheon index [20] , length ratio [21] , or straightness index [22] ) , although recent work has pointed out advantages of using the directional autocorrelation time [21 , 23] . Speed is usually quantified in terms of instantaneous speed along the trajectory or net speed over the entire assay . However , the relationship among the accuracy , persistence , and speed in chemotaxis , and whether one quantity constrains the others , is not fully understood . Are there cells that are accurate but not very persistent , or persistent but not very accurate ( Fig 1B ) ? If not , is it because such motion is possible but not fit , or is it because some aspect of cell motion fundamentally prohibits this combination of chemotactic properties ? Here we focus on how a cell’s intrinsic migration mechanism as well as properties of the external environment place constraints on its chemotactic performance . The physics of diffusion places inherent limits on a cell’s ability to sense chemical gradients [24] . These limits , along with the cell’s internal information processing and its motility mechanism , determine the accuracy , persistence , and speed of migration . Using a human breast cancer cell line ( MDA-MB-231 ) embedded within a 3D collagen matrix inside a microfluidic device imposing a chemical gradient , we are able to quantify the chemotactic performance of invasive cancer cells in response to various chemical concentration profiles . Results from chemotaxis assays are then compared with simulations and theoretical predictions in order to probe the physical limits of cancer cells to chemotaxis .
We measure accuracy using the chemotactic index ( CI ) [12–15] CI ≡ ⟨ cos θ ⟩ , ( 1 ) where θ is the angle the cell’s displacement makes with the gradient direction ( Fig 1A ) , and the average is taken over many cell trajectories . CI is bounded between −1 and 1 . For chemotaxis in response to an attractant , as in this study , CI generally falls between 0 and 1; whereas in response to a repellent , CI usually falls between −1 and 0 . CI = 1 represents perfectly accurate chemotaxis in which cell displacement is parallel to the gradient direction ( Fig 1B , top two examples ) , and CI = 0 indicates that the cells’ migration is unbiased ( Fig 1B , bottom two examples ) . The facts that CI is bounded and dimensionless make it easy to compare different values across different experimental conditions , and get an intuitive picture for the type of cell dynamics it represents . We measure persistence using the directional persistence ( DP ) , defined as the ratio of the magnitude of the cell’s displacement ( in any direction ) to the total distance traveled [20–22] ( Fig 1A ) , DP ≡ ⟨ | displacement | distance ⟩ . ( 2 ) Note that this ratio goes by several names [20–22] , and although the name we use here contains the word ‘chemotactic , ’ the ratio is in fact independent of the gradient direction . Indeed , DP measures the tendency of a cell to move in a straight line , in any direction . DP is also dimensionless and bounded between 0 and 1 , and once again intuitive sense can be made of either limit . If DP = 1 , then the cells are moving in perfectly straight lines in any arbitrary direction ( Fig 1B , right two examples ) . In contrast , a low DP is representative of a cell trajectory that starts and ends near the same location on average ( Fig 1B , left two examples ) , with DP → 0 in the limit of an infinitely long non-persistent trajectory . An alternative measure of persistence is the directional autocorrelation time τ AC = ∫ 0 ∞ d t ′ 〈 cos ( θ t + t ′ − θ t ) 〉 , where t′ is the time difference between two points in a trajectory , and the average is taken over all starting times t [21 , 23] . The advantage of the autocorrelation time is that , unlike the DP , it is largely independent of the measurement frequency and total observation time . The disadvantage is that , unlike the DP , it is not dimensionless or bounded . Although we use the DP here , we verify in S1 Fig that the autocorrelation time varies monotonically with the DP for our experimental assay . We measure speed using the instantaneous speed along the trajectory . That is , we take the distance traveled in the measurement interval Δt ( 15 minutes in the experiments , see below ) , divide it by the interval , and average this quantity over all intervals that make up the trajectory . We begin by investigating the above properties of chemotaxis in the context of metastasis , specifically the epithelial-mesenchymal transition and subsequent invasion of cancer cells . To this end , we perform experiments using a triple-negative human breast cancer cell line ( MDA-MB-231 ) . Invasion of tumor cells in vivo is aided by external cues including soluble factors that are thought to form gradients in the tumor microenvironment [2 , 5–9] . Among these soluble factors , transforming growth factor-β ( TGF-β ) is a key environmental cue for the invasion process [2 , 25–28] . Therefore , we use TGF-β as the chemoattractant . The in vivo tumor microenvironment is highly complex . As a result , in vitro platforms have been developed and widely used to investigate the cancer response to a specific cue . In this study , a microfluidic platform is used to expose the TGF-β gradient to the cells in 3D culture condition ( Fig 2A ) . The microfluidic device is designed with three different channels , a center , source , and sink channel ( Fig 2B ) . The center channel is filled with a composition of MDA-MB-231 cells and type I collagen while the medium is perfused through the side source and sink channels . TGF-β is applied only through the source channel , not the sink channel , and therefore a graded profile develops over time in the center channel by diffusion . Consequently , the MDA-MB-231 cells surrounded by type I collagen are exposed to a chemical gradient of TGF-β . To verify that a graded TGF-β profile is generated in the center channel , we utilize 10kDa FITC-dextran , whose hydrodynamic radius ( 2 . 3 nm ) is similar to that of TGF-β ( approximately 2 . 4 nm [29] ) . The fluorescence intensity is shown in Fig 2C . The profile approaches steady state within 3 hours , is approximately linear , and remains roughly stationary for more than 12 hours . Therefore , we record the MDA-MB-231 cells using time-lapse microscopy every 15 minutes from 3 to 12 hours after imposing the TGF-β . See Materials and methods for details . First , we perform a control experiment with no TGF-β to characterize the baseline of the MDA-MB-231 cell migratory behavior . Representative trajectories are shown in Fig 3A , and we see that there is no apparent preferred direction . Indeed , as seen in Fig 3C ( black ) , the CI is centered around zero , indicating no directional bias . Notably , the spread of the CI values is very broad , with many data points falling near the endpoints −1 and 1 . This is a generic feature of the CI due to its definition as a cosine: when the distribution of angles θ is uniform , the distribution of cos θ is skewed toward −1 and 1 because of the cosine’s nonlinear shape . Nonetheless , we see that the median of the CI is very near zero as expected . The speed and DP are shown in Fig 3D and 3E , respectively ( black ) . We see that the DP is significantly above zero , indicating that even in the absence of any chemoattractant , cells exhibit persistent motion . This result is consistent with previous works that showed that cells cultured in 3D tend to have directionally persistent movement unlike those in 2D [10] . Next , we expose cells to a TGF-β gradient of g = 50 nM/mm . Representative trajectories are shown in Fig 3B , and we see a possible bias in the gradient direction . Indeed , as seen in Fig 3C ( red ) , the CI is centered above zero , indicating a directional bias , and the difference with the control distribution is statistically significant ( p value < 0 . 05 ) . We also see in Fig 3D ( red ) that the speed increases , although we will see below that the increase is relatively small and that the trend is non necessarily monotonic . Finally , we see in Fig 3E ( red ) that the DP decreases , although the difference with the control is not statistically significant . These results suggest that a TGF-β gradient causes a significant increase in directional bias ( CI ) but not necessarily a significant change in cell speed or persistence ( DP ) . To confirm the trends suggested above , we evaluate the response to four different TGF-β gradient strengths , g = 0 , 1 , 5 , and 50 nM/mm , in three separate experiments each ( Fig 4A–4C; the trajectories for all experiments and g values are shown in S2 Fig ) . We see in Fig 4A that , consistent with Fig 3 , the CI is zero for the control and increases with gradient strength g . In fact , the CI appears to saturate beyond 5 nM/mm , such that its value at 50 nM/mm is not significantly larger than its value at 5 nM/mm . We also see in Fig 4B , consistent with Fig 3 , the DP slightly decreases with the gradient strength although the decrease is roughly within error bars . Finally , we see in Fig 4C that the increase in the speed is small , achieving a statistically significant difference with the control only at the largest gradient strength , and that the trend is not monotonic . A striking feature of Fig 4A is that the cells respond to a gradient as shallow as g = 5 nM/mm . To put this value in perspective , we estimate both the relative concentration change and the absolute molecule number difference across the cell body [4] . The microfluidic device is about 1 mm in the gradient direction , and therefore a cell in the middle experiences a background concentration of about c = 2 . 5 nM . Assuming the cell is on the order of a = 10 μm wide , the change in concentration across its body is ga = 0 . 05 nM , for a relative change of ga/c = 2% . The number of attractant molecules that would occupy half the cell body is on the order of ca3 = 1500 . Two percent of this is ga4 = 30 , meaning that cells experience about a thirty-molecule difference between their two halves . The same quantities are approximately ga/c = 1% and 6% , and ga4 = 60 and 300 , for amoebae in cyclic adenosine monophosphate gradients [14] and epithelial cells in epidermal growth factor gradients [30] , respectively [4] . This suggests that the response of MDA-MB-231 cells to TGF-β gradients is close to the physical detection limit for single cells . To understand the experimental observation that the CI increases with gradient strength , but the DP and speed do not ( Fig 4A–4C ) , we turn to computer simulations . The cells in the experiments are executing 3D migration through the collagen matrix ( as opposed to crawling on top of a 2D substrate ) . Nevertheless , the imaging is acquired as a 2D projection of the 3D motion . We do not expect this projection to introduce much error into the analysis because the height of the microfluidic device is less than 100 μm , whereas its width in the gradient direction is about 1 mm , and its length is several millimeters . Indeed , from the experimental trajectories ( Fig 3 ) we have estimated that if motility fluctuations in the height direction are equivalent to those in the length direction , then the error in the CI that we make by the fact that we only observe a 2D projection of cell motion is less than 1% . Consequently , for simplicity we use a 2D rather than 3D simulation of chemotaxis of a cell through an extracellular medium . Specifically , we use the cellular Potts model ( CPM ) [31 , 32] , a lattice-based simulation that has been widely used to model cell migration [33–35] ( note that whereas often the CPM is used to model collective migration , here we use it for single-cell migration ) . In the CPM , a cell is defined as a finite set of simply connected sites on a regular square lattice ( Fig 5 ) . The cell adheres to the surrounding collagen with an adhesion energy α and has a basal area A0 from which it can fluctuate at an energetic cost λ . This gives the energy function u = α L + λ ( A − A 0 ) 2 , ( 3 ) where L and A are the cell’s perimeter and area , respectively . Cell motion is a consequence of minimizing the energy u subject to thermal noise and a bias term w that incorporates the response to the gradient [33] . Specifically , for a lattice with S total sites , one update step occurs in a fixed time τ and consists of S attempts to copy a random site’s label ( cell or non-cell ) to a randomly chosen neighboring site . Each attempt is accepted with probability P = { e − ( Δ u − w ) Δ u − w > 0 1 Δ u − w ≤ 0 , ( 4 ) where Δu is the change in energy associated with the attempt . The bias term is defined as w = Δ x → · p → , ( 5 ) where Δ x → is the change in the cell’s center of mass caused by the attempt , and p → is the cell’s polarization vector ( Fig 5 inset , black arrow ) , described below . The dot product acts to bias cell motion because movement parallel to the polarization vector results in a more positive w , and thus a higher acceptance probability ( Eq 4 ) . The polarization vector is updated every time step τ according to Δ p → τ = r ( − p → + η Δ x ^ τ + ϵ q → ) . ( 6 ) The first term in Eq 6 represents exponential decay of p → at a rate r . Thus , r−1 characterizes the polarization vector’s memory timescale . The second term causes alignment of p → with Δ x ^ τ according to a strength η , where Δ x ^ τ is a unit vector pointing in the direction of the displacement of the center of mass in the previous time step τ . Thus , this term promotes persistence because it aligns p → in the cell’s previous direction of motion . The third term causes alignment of p → with q → according to a strength ϵ , where q → contains the gradient sensing information , as defined below . Thus , this term promotes bias of motion in the gradient direction . The sensing vector q → is an abstract representation of the cell’s internal gradient sensing network and is defined as q → = ⟨ ( n i − n ¯ ) r ^ i ⟩ , ( 7 ) where the average is taken over all lattice sites i that comprise the cell , and receptor saturation is incorporated as described below . The unit vector r ^ i points from the cell’s center of mass to site i , the integer ni represents the number of TGF-β molecules detected by receptors at site i , and n ¯ is the average of ni over all sites . The integer ni is the minimum of two quantities: ( i ) the number of TGF-β receptors at site i , which is sampled from a Poisson distribution whose mean is the total receptor number N divided by the number of sites; and ( ii ) the number of TGF-β molecules in the vicinity of site i , which is sampled from a Poisson distribution whose mean is ( c + gxi ) ℓ3 , where ℓ is the lattice spacing , and xi is the position of site i along the gradient direction . Taking the minimum incorporates receptor saturation , since each site cannot detect more attractant molecules than its number of receptors . The subtraction in Eq 7 makes q → a representation of adaptive gradient sensing: if receptors on one side of the cell detect molecule numbers that are higher than those on the other side , then q → will point in that direction . Adaptive sensing has been observed in the TGF-β pathway [36] in the form of fold-change detection [37] ( for shallow gradients , subtraction as in Eq 7 is similar to taking a ratio as in fold-change detection [30] ) . The simulation is performed at a fixed background concentration c and gradient g for a total time T . The position of the cell’s center of mass is recorded at time intervals Δt , from which we compute the CI , DP , and speed . The parameter values used in the simulation are listed in Table 1 and are set in the following way . The values T = 9 h , Δt = 15 min , c = 2 . 5 nM , and g = 5 nM/mm are taken from the experiments . We estimate A0 = 400 μm2 from the experiments , and we take ℓ = 2 μm , such that a cell typically comprises A0/ℓ2 = 100 lattice sites . We find that realistic cell motion is sensitive to α: when α is too small the cell is diffuse and unconnected , whereas when α is too large the cell does not move because the cost of perturbing the perimeter is too large . The crossover occurs around α ∼ ℓ−1 as expected , and therefore we set α on this order , to α = 2 μm−1 . In contrast , we find that cell motion is not sensitive to λ ( apart from λ = 0 for which the cell evaporates ) , and therefore we set λ = 0 . 01 μm−4 corresponding to typical area fluctuations of λ−1/2/A0 = 2 . 5% . In order for our Poisson sampling procedure to be valid , the time step τ must be much larger than the timescale ℓ2/D for an attractant molecule or receptor to diffuse with coefficient D across a lattice site . Taking D ∼ 10 μm2/s , we find τ ≫ 0 . 4 s . At the other end , we must have τ < Δt = 900 s for meaningful data collection . We find that within these bounds , results are not sensitive to τ , and therefore we set τ on the larger end at τ = 100 s to reduce computational run time . The parameters N , η , and ϵ are calibrated from the experimental data in Fig 4A–4C . Specifically , N sets the gradient value above which the CI saturates ( see Fig 4A ) because if the gradient is large but N is small , the cell quickly migrates into a region in which there are more attractant molecules than receptors at all lattice sites , and gradient detection is not possible . We find that N = 10 , 000 , which is a reasonable value for the number of TGF-β receptors per cell [38 , 39] , places the saturation level at roughly g = 50 nM/mm as in the experiments ( Fig 4D ) . We set ϵ = 56 μm−1 and η = 107 μm−1 to calibrate their cognate observables , CI and DP , respectively , to the corresponding experimental values at g = 5 nM/mm ( Fig 4D and 4E ) . The final parameter is the memory timescale of the polarization vector , r−1 . As seen in Fig 4E ( gray ) , we find that the behavior of the DP depends sensitively on this timescale . When r−1 is large , the DP increases with gradient strength . In contrast , when r−1 is small ( indeed , equal to the smallest timescale in the system , τ ) , the DP does not increase with gradient strength , and in fact slightly decreases ( Fig 4E , blue ) . Because the latter behavior is consistent with the experiments ( Fig 4B ) , we set r−1 = τ . We conclude that the memory timescale of MDA-MB-231 cells is very short when responding to TGF-β gradients . We validate the simulation in two ways , using the speed . First , we find that the magnitude of the speed in the simulations is on the same order as the speed in the experiments ( Fig 4C and 4F ) , i . e . , tens of microns per hour . Second , we find that the speed shows little dependence on the gradient strength in both the simulations and the experiments: it slightly increases in Fig 4C and slightly decreases in Fig 4F . Considering that the speed is not calibrated directly in our simulations , these consistencies validate the CPM as a reasonable description of the cell migration in the experiments . Our finding that the cell’s memory timescale r−1 takes its minimum value allows for the following interpretation: the parameter r couples the persistence term and the sensory term in the CPM ( Eq 6 ) . Thus , when the memory timescale r−1 is long , biased motion must be also persistent and vice versa . In contrast , when the memory timescale r−1 is short , it is possible for bias to increase without increasing persistence . Therefore , the simulations suggest that the reason that CI but not DP increases with gradient strength in the experiments , is that the drivers of sensory bias and migratory persistence in the cell’s internal network are decoupled from one another . Our finding that bias and persistence are decoupled in the simulations allows us to appeal to a much more simplified theoretical model in order to understand and predict global constraints on chemotaxis performance . Specifically , we consider the biased persistence random walk ( BPRW ) model [40 , 41] , in which bias and persistence enter as explicitly independent terms controlled by separate parameters . The BPRW has been shown to be sufficient to capture random and directional , but not periodic , behaviors of 3D cell migration [42] . Because we do not observe periodic back-and-forth motion of cells in our experiments , we propose that the BPRW is sufficient to investigate chemotactic constraints here . As in the simulations , we consider the BPRW model in 2D . In the BPRW model , a cell is idealized as a single point . Its trajectory consists of M steps whose lengths are drawn from an exponential distribution . We take M = T/Δt = 36 as in the experiments . The probability of a step making an angle θ with respect to the gradient direction is P ( θ | θ ′ ) = b cos θ ︸ bias + e p cos ( θ − θ ′ ) 2 π I 0 ( p ) ︸ persistence , ( 8 ) where θ′ is the angle corresponding to the previous step . The first term incorporates the bias , with strength b . It is maximal when the step points in the gradient direction ( θ = 0 ) and therefore promotes bias in that direction . It integrates to zero over its range ( −π < θ < π ) because the bias term only reshapes the distribution without adding or subtracting net probability . The second term incorporates the persistence , with strength p . It is a von Mises distribution ( similar to a Gaussian distribution , but normalized over the finite range −π < θ < π ) whose sharpness grows with p . It is maximal at the previous angle θ′ and therefore promotes persistence . The normalization factor I0 is the zeroth-order modified Bessel function of the first kind . The requirement that P ( θ|θ′ ) be non-negative over the entire range of θ mutually constrains b and p . However , apart from this constraint , b and p can take any positive value . We sample many pairs of b and p , reject those that violate the constraint , and compute the CI and DP from a trajectory generated by each remaining pair . The results are shown in Fig 6 ( colored circles ) . We see in Fig 6 that the BPRW model exists in a highly restricted ‘crescent’ shape within CI–DP space . As expected , the CI increases with the bias parameter b ( color of circles , from blue to red ) . The top corner corresponds to maximal bias and no persistence; indeed , when p = 0 the persistence term in Eq 8 reduces to ( 2π ) −1 , and non-negativity requires b < ( 2π ) −1 ≈ 0 . 16 , which is consistent with the upper limit of the color bar . Also as expected , the DP increases with the persistence parameter p ( size of circles , from small to large ) , although only in the lower portion where the CI is low . The crescent shape of the allowed CI and DP values in Fig 6 can be understood quantitatively because several moments of the BPRW are known analytically [41] . Specifically , the mean squared displacement and the mean displacement in the gradient direction are , in units of the mean step length , ⟨ r 2 ⟩ = 1 ( 1 − ψ ) 2 [ z 2 M ˜ 2 + 2 ( 1 − 2 z 2 − z 2 e − M ˜ ) M ˜ + 2 ( 2 z 2 − 1 ) ( 1 − e − M ˜ ) + 2 z 2 ( 1 − e − M ˜ ) 2 ] , ( 9 ) ⟨ x ⟩ = z 1 − ψ ( M ˜ − 1 + e − M ˜ ) , ( 10 ) respectively , where M ˜ = M ( 1 − ψ ) and z = χ/ ( 1 − ψ ) , with χ = ∫ − π π d θ b cos 2 θ = π b and ψ = ∫ − π π d ϕ [ 2 π I 0 ( p ) ] − 1 e p cos ϕ cos ϕ = I 1 ( p ) / I 0 ( p ) . We approximate the CI and DP in terms of these moments , CI= ⟨ x r ⟩ ≈ ⟨ x ⟩ ⟨ r ⟩ ≈ ⟨ x ⟩ ⟨ r 2 ⟩ , ( 11 ) DP= ⟨ r ⟩ M ≈ ⟨ r 2 ⟩ M , ( 12 ) and evaluate these expressions in specific limits to approximate the edges of the shape . In the limit b = 0 , Eq 11 reduces to CI = 0 ( bottom black line in Fig 6 ) . In the limit p = 0 , Eqs 11 and 12 are functions of only b and M , and b can be eliminated to yield DP = [ 1 + M ( 1 − CI 2 ) / 2 ] − 1 / 2 ( left black line in Fig 6 ) , where we have used the approximation M ≫ 1 ( see Materials and methods ) . Note here that when CI = 0 we have DP ≈ ( M/2 ) −1/2 for large M , which makes sense because for a simple random walk ( p = b = 0 ) the displacement goes like M1/2 while the distance goes like M , such that DP ∼ M−1/2 . Finally , the right edge corresponds to the maximal value of p for a given b , for which we compute the approximation curve parametrically ( right black line in Fig 6; see Materials and methods ) . We see in Fig 6 that these approximate expressions slightly underestimate the CI and overestimate the DP , but otherwise capture the crescent shape well . The under- and overestimation are due to the approximation 〈 r 〉 ≈ 〈 r 2 〉 in Eqs 11 and 12: because σ r 2 = 〈 r 2 〉 − 〈 r 〉 2 ≥ 0 for any statistical quantity , we have 〈 r 2 〉 ≥ 〈 r 〉 , making Eq 11 an underestimate and Eq 12 an overestimate . The crescent shape can also be understood intuitively . First , we see that the DP cannot be smaller than a minimum value ( region I in Fig 6 ) . This is because the trajectory length M is finite , and as discussed above , the DP only vanishes for infinitely long trajectories . If M were to increase , the crescent would extend further toward DP = 0 . Second , we see that the top of the crescent bends away from the CI →1 , DP →0 corner ( region II in Fig 6 ) . In other words , it is not possible to have high bias without any persistence . This is because if the bias is strong , then cells will track the gradient very well . Consequently , they will move in nearly straight lines in the gradient direction , and straight movement corresponds to high persistence . This is a bias-induced persistence , distinct from the bias-independent persistence in the lower-right corner of the crescent . Finally , we see that the bending shape of the crescent implies that no solutions exist at large DP and intermediate CI ( region III in Fig 6 ) . In other words , it is not possible to have high persistence with partial bias . This is because , as mentioned above , persistence is induced either ( i ) directly , as a result of a large persistence parameter p which is independent of the bias , in which case the CI is low; or ( ii ) indirectly , as a result of a large bias parameter b , in which case the CI is high . Neither of these mechanisms permits intermediate bias , and therefore high persistence can be accompanied only by low or high directionality . Together , these features of the crescent shape imply that specific modes of chemotaxis are prohibited under our simple model , as indicated by the regions I , II , and III . Finally , the crescent shape provides a qualitative rationale for the data from the simulations and experiments , which are overlaid in the cyan and red squares in Fig 6 , respectively . Specifically , the shape of the crescent is such that if a cell has a low CI and intermediate DP ( bottom right corner of the crescent ) and its CI increases , its DP must decrease ( solid magenta arrow in Fig 6 ) . In contrast , a simultaneous increase in CI and DP from this starting position is not possible according to the model ( dashed magenta arrow in Fig 6 ) . We see that the data are qualitatively consistent with this predicted trend , as an increase in the CI corresponds to a decrease in the DP in both the experiments and the simulations ( Fig 6 , squares ) . There is quantitative disagreement , in the sense that the data do not quite overlap with the crescent , but this is a reflection of the extreme simplicity of the BPRW model . Nonetheless , the qualitative features of the BPRW model are sufficient to explain the way in which accuracy and persistence are mutually constrained during the chemotaxis response of these cells .
By integrating experiments with theory and simulations , we have investigated mutual constraints on the accuracy ( CI ) , persistence ( DP ) , and speed of cancer cell motion in response to a chemical attractant . We have found that while the CI of breast cancer cells increases with the strength of a TGF-β gradient , the speed does not show a strong trend , and the DP slightly decreases . The simulations suggest that the decrease in DP is due to a decoupling between sensing and persistence in the migration dynamics . The theory confirms that the decrease in DP is due to a mutual constraint on accuracy and persistence for this type of decoupled dynamics , and more generally , it suggests that entire regions of the accuracy–persistence space are prohibited . The present results provide some insights into TGF-β induced migration mechanisms . Multiple signaling pathways induced by TGF-β affect the dynamics of actin polymerization regulating cell migratory behaviors [27 , 43–45] . Among these , phosphatidylinositol 3-kinase ( PI3K ) and the small GTPase-Rac1 signaling have been reported to promote actin organization of breast cancer cells in response to TGF-β [45 , 46] . PI3K and the Rho-family GTPase networks ( including Rac1 , RhoA and Cdc42 ) have been widely studied in chemotaxis , which regulates cell polarity and directional sensing [47–50] . The PI3K activity , thus , can possibly explain the present chemotactic responses of the breast cancer cells to TGF-β gradient . Recent studies have shown that PI3K is relevant to the accuracy of the cell movement in shallow chemoattractants , whereas it does not induce the orientation of cell movement in steep gradients; rather , PI3K contributes the motility enhancement [51 , 52] . These results can be correlated with the cell motility trend in the present experimental results . In addition , the PI3K signaling pathway has been reported not to mediate the persistence of cell protrusions which could be directly related to the DP [47 , 48] . The directional persistence could be more relevant to the polarity stability which is hardly controlled by chemotaxis [47] as presented in the present results . In TGF-β molecular cascades , activation of SMAD proteins could also affect the actin dynamics . Since SMAD-cascades include negative feedback inhibiting Rho activity [43 , 44] , it may affect the cell responses highly promoted in CI but not in speed . However , the underlying molecular mechanisms need further research . Our finding that sensing and persistence are largely decoupled in the migration dynamics is related to the view that directional sensing and polarity are separate but connected modules in chemotaxis [11] . Indeed , CI , DP , and speed in our study play the roles of the directional sensing , polarity , and motility modules , respectively , that have been shown to reproduce many of the observed behaviors of chemotaxing cells . Moreover , several of the the molecular signaling pathways discussed above , including those involving PI3K and Rho family GTPases , have been proposed as the potential networks corresponding to these modules [11] . Several predictions arise from our work that would be interesting to test in future experiments . First , our simulation scheme assumes that the saturation of the CI with gradient strength ( Fig 4A ) is due to limited receptor numbers . However , alternative explanations exist that are independent of the receptors , such as the fact that it is more difficult to detect a concentration difference on top of a large concentration background than on top of a small concentration background due to intrinsic fluctuations in molecule number [30 , 53] . An interesting consequence of our mechanism of receptor saturation is that , at very large gradients ( beyond those of Fig 4A ) , the CI would actually decrease because all receptors would be bound . It would be interesting to test this prediction in future experiments . Second , our work suggests that not all quadrants of the accuracy–persistence plane are possible for cells to achieve ( Fig 6 ) . It would be interesting to measure the CI and DP of other cell types , in other chemical or mechanical environments , to see if the crescent shape seen in Fig 6 is a universal restriction , or if not , what new features of chemotaxis are therefore not captured by the modeling . In this respect , the work here can be seen as a null model , deviations from which would indicate new and unique types of cell motion .
Human breast adenocarcinoma cells ( MDA-MB-231 ) were cultured in Dulbecco’s Modified Eagle Medium/Ham’s F-12 ( Advanced DMEM/F-12 , Lifetechnologies , CA , USA ) supplemented by 5% v/v fetal bovin serum ( FBS ) , 2 mM L-glutamine ( L-glu ) , and 100 μg ml-1 penicillin/streptomycin ( P/S ) for less than 15 passages . MDA-MB-231 cells were regularly harvested by 0 . 05% trypsin and 0 . 53mM EDTA ( Lifetechnologies , CA , USA ) when grown up to around 80% confluency in 75 cm2 T-flasks at 37 °C with 5% CO2 incubation . Harvested cells were used for experiments or sub-cultured . Cell-matrix composition was prepared in the microfluidic device . For the composition , MDA-MB-231 cells were mixed with 2 mg/ml of type I collagen ( Corning Inc . , NY , USA ) mixture prepared with 10X PBS , NaOH , HEPE solution , FBS , Glu , P/S , and cell-culture level distilled water after centrifuged with 1000 rpm for 3 minutes . The cell mixture was filled in center-channel of the microfluidic devices and incubated in at 37 °C with 5% CO2 . The cells in the collagen matrix were initially cultured in basic medium ( DMEM/F12 supplemented by 5% v/v FBS , 2 mM L-glu , and 100 μg ml−1 p/s ) for 24 hours . Then the cells were exposed by reduced serum medium for another 24 hours , which was advanced DMEM/F12 containing 1% v/v FBS , 2 mM L-glu , and 100 μg ml−1 p/s [54] . After 24 hour-serum starvations , cells were exposed by a gradient of transforming growth factor beta-1 ( TGF-β1 , Invitrogen , CA , USA ) . The microfluidic device was designed to generate a linear gradient of soluble factors ( Fig 2 ) . The device is composed of three channels which are 100 μm in thickness as described previously [55] . A center channel that is 1 mm wide aims to culture tumor cells with ECM components . The center channel is connected to two side channels . The 300 μm-wide side channels are connected to large reservoirs at the end ports including culture medium . Since the side channels are in contact with the top and bottom sides of the center channel , the growth factor gradient can be generated by diffusing the soluble factor from one of the side channels , a source channel , to the other , a sink channel . Assuming there is neither pressure difference nor flow between the side channels , the concentration of a given factor can be described by the chemical species conservation equation as follows: ∂ c i ∂ t = D i · ∇ c i ( 13 ) Once the concentration profile in the center channel reaches steady state , the linear profile persists for a while and can therefore be approximated by assuming the boundary conditions of concentration at the side channels are constants . To verify the diffusion behavior , the gradient formation was examined by using 10k Da FITC-fluorescence conjugated dextran ( FITC-dextran ) . FITC-dextran solution was applied in the source channel while the sink channel was filled with normal culture medium . The FITC-dextran concentration profile was evaluated by the FITC fluorescent intensity in the center channel . To disregard the effect of photo-bleaching on the results , the intensity was normalized by the intensity of the source channel . The normalized intensity was reasonably considered since the fluorescence intensity of the source channel consistently remained as maximum due to the large reservoirs . The FITC dextran intensity profile ( Fig 2C ) showed that the linear profile was developed within 3 hours after applying the source and continued for more than 9 hours . Cell behaviors were captured every 15 minutes for 9 hours using an inverted microscope ( Olympus IX71 , Japan ) equipped with a stage top incubator as described previously [56–58] , so that the microfluidic platform could be maintained at 37 °C in a 5% CO2 environment during imaging . The time-lapse imaging was started 3 hours after applying TGF-β1 solution in the source channel to have sufficient adjusting time . To analyze each cell behavior , a cell area in the bright field images were defined by a contrast difference between the cells and a background , and the images were converted to monochrome images by using ImageJ . Cell trajectories were demonstrated by tracking centroids of the cell area . In tracking the cell movements , cells undergoing division were excluded to avoid extra influences to affect cell polarity [59] . Moreover , stationary cells due to the presence of the matrix were excluded [26 , 59–61] . The stationary cells were defined as the cells that moved less than their diameter . A migration trajectory was defined by connecting the centroids of a cell from each time point . In examining the chemotactic characteristics of each group , more than 40 cell trajectories were evaluated per a group . A data point in Fig 3C–3E indicates each metric of a cell trajectory showing distribution characteristics with a box plot . The box plot includes boundaries as quadrants and a center as a median . The distribution of each metric was statistically analyzed by using Mann-Whitney U-test . This non-parametric method was used since the distribution was not consistently normal ( the CI is a function of cosine ) . The significant change on the population lies on the biased distribution of each cell parameter when the p value < 0 . 05 . Furthermore , the experiments were repeated at least 3 times and reported with means of medians ± standard estimated error ( S . E . M . ) in Fig 4A–4C . To evaluate physical limits on each metric , the data points were compared each other using a student t-test . The statistical significance between comparisons were examined when the p value < 0 . 05 . In the limit p = 0 , Eqs 9 and 10 become ⟨ r 2 ⟩ = z 2 M 2 + 2 ( 1 − 2 z 2 ) M + 2 ( 3 z 2 − 1 ) , ( 14 ) ⟨ x ⟩ 2 = z 2 ( M − 1 ) 2 , ( 15 ) where z = πb , and we have neglected the exponential terms in the limit M ≫ 1 . Defining the small parameter ϵ = 1/M , these expressions become ⟨ r 2 ⟩ = z 2 M 2 ( 1 + c ϵ ) , ( 16 ) ⟨ x ⟩ 2 = z 2 M 2 ( 1 − 2 ϵ ) ( 17 ) to first order in ϵ , where c ≡ 2 ( z−2 − 2 ) . Inserting these expressions into Eqs 11 and 12 , we obtain CI 2 = 1 − ( c + 2 ) ϵ , ( 18 ) DP 2 = z 2 ( 1 + c ϵ ) ( 19 ) to first order in ϵ . Because z and c are both functions only of b , we eliminate b from Eqs 18 and 19 to obtain CI 2 = 1 − 2 ϵ 1 − DP 2 DP 2 ( 20 ) to first order in ϵ . This expression is equivalent to that given below Eq 12 and provides the left black line in Fig 6 . The right black line in Fig 6 corresponds to the maximal value of p for a given b that keeps Eq 8 non-negative . Non-negativity requires that the sum of the minimal values of each term in Eq 8 is zero: −b + e−p/[2πI0 ( p ) ] = 0 . With this expression for b in terms of p , Eqs 11 and 12 become functions of only p and M . Therefore , by varying p , we compute the right black line parametrically . | One of the most ubiquitous and important cell behaviors is chemotaxis: the ability to move in the direction of a chemical gradient . Due to its importance , key aspects of chemotaxis have been quantified for a variety of cells , including the accuracy , persistence , and speed of cell motion . However , whether these aspects are mutually constrained is poorly understood . Can a cell be accurate but not persistent , or vice versa ? Here we use theory , simulations , and experiments on cancer cells to uncover mutual constraints on the properties of chemotaxis . Our results suggest that accuracy and persistence are mutually constrained . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] | [] | 2019 | Physical constraints on accuracy and persistence during breast cancer cell chemotaxis |
Placozoans are a phylum of nonbilaterian marine animals currently represented by a single described species , Trichoplax adhaerens , Schulze 1883 . Placozoans arguably show the simplest animal morphology , which is identical among isolates collected worldwide , despite an apparently sizeable genetic diversity within the phylum . Here , we use a comparative genomics approach for a deeper appreciation of the structure and causes of the deeply diverging lineages in the Placozoa . We generated a high-quality draft genome of the genetic lineage H13 isolated from Hong Kong and compared it to the distantly related T . adhaerens . We uncovered substantial structural differences between the two genomes that point to a deep genomic separation and provide support that adaptation by gene duplication is likely a crucial mechanism in placozoan speciation . We further provide genetic evidence for reproductively isolated species and suggest a genus-level difference of H13 to T . adhaerens , justifying the designation of H13 as a new species , Hoilungia hongkongensis nov . gen . , nov . spec . , now the second described placozoan species and the first in a new genus . Our multilevel comparative genomics approach is , therefore , likely to prove valuable for species distinctions in other cryptic microscopic animal groups that lack diagnostic morphological characters , such as some nematodes , copepods , rotifers , or mites .
Placozoans Grell , 1971 , are small , benthic marine animals found worldwide in various habitats [1–6] . To date , only a single species has been described , Trichoplax adhaerens Schulze 1883 . Animals are flat and have a typically disc-like morphology but have the capacity to change shape [7–9] . The lack of symmetry axes , neurons , and defined muscle cells , and the presence of only six morphologically distinguishable somatic cell types ( [9 , 10]; Fig 1 , S1 Fig ) , makes the Placozoa morphologically the most simply organized animals . The prominent placozoan modes of reproduction are asexual , i . e . , binary fission and budding [8 , 9 , 11–13] that produce genetically identical clones . Sexual reproduction has rarely been observed under laboratory condition [14–19] , but both oocytes and sperm cells have been reported [14 , 17 , 19] , and fertilization , likely coupled with genetic exchange , was confirmed based on structural similarities of the placozoan eggshell with the fertilization membrane of other animal groups [16] . No sexually reproducing individual has ever been reported from the wild . However , the occurrence and success of sexual reproduction in the field have been demonstrated by DNA sequence analyses , since nuclear-encoded marker genes have revealed the occurrence of allele sharing and mixing of heterozygous alleles in a natural placozoan population isolated from a Caribbean habitat [20] . These molecular signatures for genetic exchange prove that sexual reproduction does occur and that the life cycle is completed in the natural environment . However , all efforts to follow the placozoan embryonic development in the laboratory have failed to date . All embryos died at an early stage during development , never reaching beyond the 128-cell stage [19] . The fragmentation of the nucleus in the zygote [21] was previously suggested as the reason for the termination of development , although this has been questioned [19] . This ambiguity and scarcity of information has , therefore , left us with a large knowledge gap regarding the life history of the Placozoa and has resulted in speculations of the existence of a missing life stage ( compare [22] ) . The genome of the diploid T . adhaerens was sequenced previously [22] , revealing that this morphologically very simple animal harbors a rich repertoire of gene families [22] . These families are known from bilaterian animals and are typically associated with a considerable cell type diversity , a complex body plan , developmental processes , and behavioral responses to external stimuli [10 , 23–31] . Additionally , single-gene molecular phylogenetics have identified a sizeable cryptic diversity within placozoans collected worldwide; but while their gross morphology is highly plastic , morphologically , all isolates fit the description of T . adhaerens [32] ( Fig 1 ) . The high intraspecific shape variability , coupled with an ultraconserved internal structure ( Fig 1 , S1 Fig ) , does not allow the establishment of reliable diagnostic morphological characters in the Placozoa , hindering attempts to characterize their diversity . While these single-marker studies provided clear indications that additional species may be uncovered in the Placozoa , two fundamental questions remain: how different are placozoans at the nuclear genome level , and what can we learn from comparative genomics about the evolution and diversity of placozoans ? To address these questions , we generated a high-quality draft genome of a placozoan lineage that is genetically distantly related to T . adhaerens [3 , 5] and performed a multilevel comparison , including genome synteny , gene clustering , gene ontology ( GO ) term enrichment , allele sharing , and cross-phylum comparative distance analyses . This approach , together with the morphological characterization of the lineage H13 , allowed us to assign a taxonomic status to morphologically cryptic taxa and led to the establishment of the second placozoan species in a new placozoan genus .
Based on mitochondrial 16S ribosomal DNA ( rDNA ) analyses , the genetic lineage H13 is among the most distantly related haplotype to T . adhaerens ( lineage H1 ) [5] , whose nuclear genome has been sequenced previously [22] . We hypothesized that the substantial 16S rDNA divergence might also be reflected on the whole-genome scale and , therefore , targeted H13 for nuclear genome sequencing . To assemble the genome of H13—a new species described here , called H . hongkongensis nov . gen . , nov . spec . ( Fig 1 , S1 Fig; see species description in Material and methods; Tables 1 and 2 ) —we generated 24 Gb of paired-end reads and 320 Mb of Moleculo ( Illumina Artificial Long Synthetic ) reads . Our final , highly complete 87-megabase assembly contained 669 high-quality and contamination-filtered contigs with an N50 of 407 kb ( S1 Table; S2–S4 Figs ) , 7 megabases smaller than the T . adhaerens contig assembly . The overall calculated genome heterozygosity ( based on single-nucleotide polymorphism [SNP] counts , see S2 Table ) was 1 . 6% , which is moderate for a marine animal but about average when compared to arthropods and high in comparison to terrestrial chordates [33] . This value cannot be compared to T . adhaerens because of the low genome coverage of the latter , which does not allow haplotype phasing . We annotated the genome with a combination of 15 . 3 Gb of RNA-Seq and ab initio methods to yield 12 , 010 genes ( S1 Table , S1 & S2 Data ) . A high percentage of raw reads mapped back to the genome ( S3 Table ) , and between 90 . 8%–95 . 3% of the 978 genes in the BUSCO v3 Metazoa dataset were identified in the transcriptome and the ab initio gene models , respectively ( S4 Table ) . Together , this suggests an almost complete assembly and annotation , in which 96 . 5% of the genes in the H . hongkongensis genome were expressed in what are commonly considered adult animals . In our gene set , H . hongkongensis had 490 more genes than the 11 , 520 genes reported in the original T . adhaerens annotation from 2008 [22] . We reannotated T . adhaerens with AUGUSTUS and found an additional 1 , 001 proteins and also managed to complete formerly partial proteins ( for T . adhaerens Blast2GO protein annotations see S3 Data ) . This approach added 4 . 4 Mb of exons to the T . adhaerens annotation , an increase of 28% of exonic base pairs to the original annotation . The new T . adhaerens annotation now has 511 more genes than H . hongkongensis , which accounts for some portion of the size difference between the two genomes . Moleculo reads also enabled us to assemble very large reference contigs , the longest being over 2 Mb . We compared the organization of genes in H . hongkongensis to the 10 longest scaffolds in the T . adhaerens genome ( size range 2 . 4–13 . 2 Mb; accounting for 66% of the T . adhaerens assembly ) . We found 144 contigs >100 kb from H . hongkongensis that aligned to these 10 scaffolds , accounting for 69% of the H . hongkongensis assembly ( Fig 2A ) . Mean gene collinearity ( i . e . , the same genes in the same direction ) in this reduced genome representation was in the range of 69 . 5% to 78 . 8% ( mean 73 . 6% ± 5 . 5%; see S5 Table ) . The mean number of genes per syntenic block was 33 . 8 ( ±25 . 2 ) in the reduced set and 33 . 9 ( ±24 . 7 ) when comparing full genomes ( S5 Fig ) , which indicates that the reduced set is representative for both complete genomes . Although much of the gene order is conserved between the two species , we counted 2 , 101 genes ( out of the 8 , 260 genes in the 10 scaffolds ) that were inverted or translocated within the same scaffold relative to the order in the T . adhaerens scaffolds . These numbers seem low when compared to the fast-evolving bilaterian genus Drosophila [34 , 35] or Caenorhabditis [36] , but they are in the range of rearrangements found between mouse and human [37] . Comparison to Bilateria , however , might be misleading ( see also results on genetic distances below ) , and genome rearrangement events might be more favored in some bilaterian taxa because of inherent genomic traits such as transposon-induced rearrangement hotspots [38] . Nonetheless , the high percentage of rearrangements between T . adhaerens and H . hongkongensis is clear evidence for a deep genetic separation of both lineages . To estimate how divergent the two placozoan genomes are at the sequence level , we calculated genetic distances for 6 , 554 one-to-one orthologs . Between H . hongkongensis and T . adhaerens , genetic distances ranged from 0 . 9% to 80 . 1% ( mean 28 . 3% ± 12 . 9% ) for proteins and 7 . 4% to 80 . 7% ( mean 28 . 5% ± 9 . 9% ) for coding sequences ( CDSs ) , respectively ( Fig 2B ) . To assess if certain genes are under positive ( diversifying ) selection , indicative of functional evolution , we calculated the ratio of nonsynonymous to synonymous nucleotide substitutions ( dN/dS ratio [39] ) for each H . hongkongensis and T . adhaerens one-to-one ortholog pair . Results show that most orthologs ( 97% ) are under strong purifying selection ( dN/dS < 0 . 5 ) . One might hypothesize that strong purifying selection pressure is the reason for the phenotypic stasis we see in modern placozoans . However , more placozoan genomes in the phylum are clearly needed to test this hypothesis . Despite this strong tendency toward purifying selection , a high proportion of orthologs ( 46% ) showed larger protein distance than CDS distance and , therefore , an accumulation of double or triple mutations in already mutated codons , which led to amino acid substitutions ( S6 Fig ) . Only 3 of the 6 , 554 one-to-one orthologs had dN/dS ratios slightly >1 , indicating positive selection ( S7 Data; see S6 Fig for an estimate of mutation saturation in codons ) . One of these seems placozoan specific , since it could not be annotated because of missing UniProt BLAST hits and InterPro domains , respectively . For the second , GO annotation and InterPro IDs indicate a role in telomere maintenance . The third positively selected gene ( CYP11A1 ) is putatively a cholesterol side-chain cleavage enzyme acting in the mitochondrion . The roughly 4x coverage of the genome with long Moleculo reads ( N50 of 5 . 4 kb ) allowed the assembly of large haplocontigs ( i . e . , contigs representing both haplotypes of the genome ) . This phasing information for large parts of the genome facilitated the isolation of 2 , 870 one-to-one orthologs with both full-length alleles after a highly stringent filtering procedure . Only by using the phasing information we were able to show that many orthologs with high allelic variation in H . hongkongensis were also profoundly different between the species ( S7 Fig ) . This indicates that genetic sequence adaptation already takes place at the population level and is further magnified between species in the same genes . The Markov cluster ( MCL ) analysis identified 6 , 644 true one-to-one orthologs ( for an overview of ortholog categories , see Material and methods and [40] ) for both placozoan species ( 55% of all proteins in H . hongkongensis and 53% in T . adhaerens , respectively ) ( S8 Fig ) . A fraction of 465 ( 3 . 8% ) H . hongkongensis and 1 , 036 ( 8 . 3% ) T . adhaerens proteins , respectively , did not have reciprocal BLAST hits . The difference in the non-BLAST hits almost perfectly matches the differences in total gene numbers , which is probably an indication that genes without a homolog in H . hongkongensis account at least partially for the slightly higher gene number in T . adhaerens . A high proportion of proteins had BLAST hits to the UniProt database , and only 15 . 4% ( 1 , 859 ) and 19 . 0% ( 2 , 384 ) of H . hongkongensis and T . adhaerens proteins , respectively , did not have BLAST hits to metazoans included in UniProt . Placozoan-specific duplications constitute a significant proportion of both proteomes , with 3 , 943 ( 32 . 8% ) co-orthologs in H . hongkongensis and 3 , 484 ( 27 . 8% ) in T . adhaerens . The enrichment analyses for the proteins in each non-BLAST-hit bin identified unique GO terms in all three GO categories among the first five most significantly enriched GO terms ( S4 & S5 Data ) . The same applies to one-to-many and many-to-one co-orthologs in both species . The enrichment analyses further indicate that both placozoan species have multiple co-orthologs associated with G-protein-coupled receptor ( GPCR ) signaling . A rich repertoire of GPCRs has been identified in T . adhaerens [22] , but here , we were able to identify independent GPCR duplications in H . hongkongensis and T . adhaerens , respectively ( S6 Data ) . Furthermore , we identified multiple enriched GO terms related to synaptic activity in all co-ortholog categories ( S5 Data ) and both placozoan species . This points to a plethora of independent duplication events in gene families related to sensory capacities . Despite lacking neurons ( based on traditional morphological classifications ) , T . adhaerens has previously been shown to stain positive for FMRFamide [10 , 41] and recently even to change behavior when exposed to physiologically relevant levels of neuropeptides [31] . Based on the identification of vast and independent gene family expansions in both placozoans , we propose that adaptation in the Placozoa , ultimately leading to speciation , is coupled with independent gene duplications as suggested , for example , for bacteria , yeast , plants , and other animals ( compare [42–45] ) . H . hongkongensis was isolated from a stream running through a mangrove with rapid drops in salinity and temperature , especially during heavy rainfall in the summer . We hypothesize that the presence of multiple divergent copies of genes involved in various processes , such as behavior and metabolism ( compare [42 , 43] ) , in addition to a situation-dependent expressional fine-tuning of these copies was necessary for adaptation to this habitat and would facilitate speciation . We furthermore propose that the presence of multiple copies of genes and their expression does not affect the phenotype but instead provides a genetic toolkit for gradual physiological responses to ( changes in ) the environment . All internal Linnaean ranks within the Placozoa are , as yet , undefined [5] . Despite efforts to identify them , reliable diagnostic morphological characters , commonly used for defining animal species , are lacking in the Placozoa [46] . Thus , all present taxonomic definitions in the phylum must solely rely on diagnostic molecular characters . In other taxonomic groups ( e . g . , bacteria and archaea [47] , protists [48 , 49] , and fungi [50] ) , purely sequence-based approaches and working models for the distinction of taxa have been proposed and are generally well established and widely accepted [51] . In animals , such methods ( which may be based on distances , on trees , or on allele sharing; [52] ) are currently under development and have been used in rare cases to identify and describe cryptic species [53] . In a first step to converting the identified genomic differences into a taxonomically meaningful system , we studied reproductive isolation by addressing allele sharing within placozoan isolates from different localities . To identify reproductive isolation , a conspecificity matrix ( CM ) was generated [54] . The CM was based on three nuclear genes encoding ribosomal proteins and clearly identified reproductive isolation between placozoan clades ( Fig 3 ) . This approach extends a previous study that has uncovered sexual reproduction only within one placozoan haplotype ( H8 ) [20] and provides clear evidence that the previously established placozoan clades ( based on 16S genotyping ) are reproductively isolated biological species . We have shown that biological species exist in the Placozoa . Previous studies have furthermore provided first indications for the existence of deeper differences between placozoan lineages [1 , 3] , with as-yet-unknown correspondence to , for example , the Linnaean ranks of genus , family , order , and class . However , these observed deeper divergences were based on single marker genes only , and no diagnostic morphological traits could be identified to establish a firm , higher-level , systematic framework in the Placozoa . To further estimate the level of taxonomic relatedness between T . adhaerens and the new placozoan species H . hongkongensis ( strain H13 ) , and in an attempt to initiate a higher-level taxonomic system for the Placozoa , we performed cross-phylum multimarker sequence divergence analyses . To do so , we compared the variation between the two placozoans to variation within the other three nonbilaterian phyla , Cnidaria , Ctenophora , and Porifera ( compare [1] ) , as well as the bilaterian phylum Chordata . Marker sets included a nuclear protein set of 212 concatenated proteins ( dataset 1 , a taxon-extended matrix from [56]; S7–S9 Tables; see Fig 4 ) as well as 5 selected genes with different substitution rates ( S9–S14 Figs ) , all commonly used for DNA barcoding and molecular systematics . Across individual markers , it appears that the phylogenetic ranks are most robust in the Cnidaria , in which the partitioning of molecular variation matches the established taxonomy , in that Linnaean ranks consistently correspond to the greater distance between groups ( Fig 4; S9–S14 Figs ) . The same is true for the Chordata , which was included in our distance calculations for the 212 nuclear protein set as an example of a bilaterian phylum with a high taxonomic coverage ( many genomes are available for this group ) . However , distances in chordates are , in general , much lower when compared to the overall more similar nonbilaterian phyla . This indicates that ( i ) genetic distances and corresponding Linnaean rank assignments in Chordata cannot be compared to nonbilaterian lineages and ( ii ) that comparisons among nonbilaterians are better suited to guide taxonomic ranking of the two placozoan species . We consequently used genetic distances in the Cnidaria as an approximation and comparative guideline for the higher systematic categorization of the new placozoan species . Genetic distances between H . hongkongensis and T . adhaerens were higher than those for the Cnidaria in five of the six marker sets at the generic level but lower at the family level for all markers ( S14 Fig , S10 Table ) , which , cautiously interpreted , supports genus-level genetic differences between the two placozoans . A clear split of the Placozoa in the molecular groups “A” and “B” was previously shown by the rearrangement pattern of mitochondrial genomes [61] and compensatory base changes in the internal transcribed spacer 2 ( ITS2 ) [55] . The conspecificity analysis , the high amount of genomic rearrangement , and the large-scale independent gene duplication history , as well as the genetic distances in six independent datasets , strongly support this split ( Fig 3 ) . Since clades were identified as the primary taxonomic units—i . e . , biological species—these two previously identified higher-level placozoan “groups” consequently represent at least the genus level in the Linnaean hierarchical system . We therefore establish the new genus Hoilungia for the former group “A” ( clades III–VII ) , which is , so far , the single sister genus to Trichoplax ( former group “B”; clades I and II ) . Future research efforts focusing on genome sequencing of additional placozoan clades/species will likely help to establish a broader and more detailed systematic framework for the Placozoa and provide further insights into the mechanisms and driving forces of speciation in this enigmatic marine phylum . Recent discussions about the phylogenetic position of placozoans have largely been based on the T . adhaerens genome . A better sampling of placozoan genomic diversity is , however , needed [62] to address their placement in the metazoan tree of life . In this context , it is important to first assess if adding another placozoan genus would break up the long placozoan branch . The inclusion of a single representative of a clade with a very long terminal branch , or fast-evolving taxa that can have random amino acid sequence similarities , may result in erroneous groupings in a phylogeny ( so-called “long-branch attraction artefacts” ) [63 , 64] . To address these questions , we generated a highly ( taxa ) condensed version of the full protein matrix from Cannon and colleagues [56] ( termed dataset 2; with less than 11% missing characters and 194 genes ) . We additionally created a Dayhoff 6-state recoded matrix [65 , 66] of this second set to reduce amino acid compositional heterogeneity , which is also known to be a source of phylogenetic error [67 , 68] . Phylogenetic analyses were performed on these two matrices ( protein and Dayhoff-6 recoded ) , using the site-heterogeneous CAT-GTR model in PhyloBayes-MPI [69] and using the site-homogenous GTR model both in Phylobayes-MPI and RAxML ( RAxML , protein only ) [70] , as well as the LG model in RAxML ( protein only ) . The resulting trees ( S15–S20 Figs ) of the highly dense gene matrix ( S21 Fig ) suggest a sister group relationship of the Placozoa to a Cnidaria + Bilateria clade with both CAT-GTR ( Protein , Dayhoff-6 recoded , S15–S17 Figs ) and GTR models ( Protein , S18 Fig ) in PhyloBayes , or these relationships are unresolved ( RAxML , protein , both GTR , S19 Fig , and LG , S20 Fig ) . This is in agreement with some previous findings [56 , 64 , 71–74] and with recent studies using a large gene set and intense quality controls [64] as well as improved modeling of compositional heterogeneity [68] . In addition , the sister group relationship of the Placozoa to the Cnidaria + Bilateria clade is corroborated by independent data—namely , the analysis of metazoan genome gene content [73 , 75 , 76] .
Phylum: Placozoa , Grell 1971 [77] Type Family: Trichoplacidae , Bütschli and Hatschek 1905 [78] , synonymized with “Trichoplaciden” ( in German original ) , Haeckel 1896 [79] . Diagnosis: We assign all currently known 19 placozoan genetic lineages ( 16S haplotypes H1-H19; [5] ) to the Trichoplacidae . The description of T . adhaerens Schulze 1883 applies to all . Type Genus: Hoilungia , nov . gen . , Eitel , Schierwater , and Wörheide Hoilungia is the second genus of the family Trichoplacidae . Etymology: Hoilungia , pseudo-Latinized from “Hoi Lung , ” Cantonese , meaning “sea dragon , ” which is based on the shape-shifting dragon king in Chinese mythology . Diagnosis: Gross and fine morphology appear similar among all placozoans studied to date . We therefore use molecular diagnostics to define Linnaean ranks . Among all tested markers , the mitochondrial large ribosomal subunit 16S rDNA appears to be the most variable among placozoans and other nonbilaterian phyla , and the mean pairwise distance is closest to that calculated for the nuclear dataset in most cases ( S14 Fig ) . This marker also best mirrored classical taxonomy in the Porifera and Cnidaria ( S11 Fig; in Ctenophora , 16S rDNA is highly derived and hard to identify [80] ) . According to these data , molecular diagnostics based on differences in the 16S rDNA appear to be suitable for current and future designation of species in the Placozoa , which is in agreement with previous results [3] . Diagnostics are here , therefore , defined by nucleotide substitutions in the 16S rDNA . Full-length 16S rDNA sequences of T . adhaerens and H . hongkongensis ( clonal strain “M2RS3-2” ) , as well as for the undescribed Placozoa sp . H4 and sp . H8 , were aligned with MAFFT v7 . 273 [81] using the GINSI option and otherwise default settings . Ambiguously aligned 5′ and 3′ sequence ends were removed . To this alignment , we added all currently available placozoan 16S haplotype sequences [5] using MAFFT [added option:—add] . The final alignment contained all 19 placozoan haplotypes and had a length of 2 , 551 nucleotides ( including gaps ) . The region for identification of diagnostic nucleotides was restricted to a part of the 16S alignment that was previously shown to be suitable and sufficient for molecular haplotype discrimination [1 , 3 , 5] . We furthermore restricted the identification of diagnostics to stem regions of this rDNA to omit uncertainties in future taxonomic assignment due to ambiguously aligned loop regions . To identify molecular diagnostics for the genus Hoilungia , we screened for molecular synapomorphies ( nucleotide exchanges ) within the placozoan 16S group “A” ( clades III–VII; [5 , 61] ) versus group “B” ( clades I and II ) . Molecular diagnostics for Hoilungia and Trichoplax are summarized in Table 1 . Type species: H . hongkongensis , nov . spec . , Eitel , Schierwater , and Wörheide . Diagnosis: To identify molecular species diagnostics , we determined unique substitutions ( based on the alignment used for genus diagnostics before ) for H . hongkongensis ( clade V ) in comparison to the other Hoilungia clades ( III , IV , VI , and VII ) . Molecular diagnostics for H . hongkongensis are summarized in Table 2 . Type locality: A single specimen of H . hongkongensis ( clonal strain “M2RS3-2” ) was isolated in the Ho Chung River close to a small mangrove at Heung Chung village , Hong Kong ( 22 . 352728N 114 . 251733E ) , on June 6 , 2012 . Type specimen: One specimen of H . hongkongensis ( clonal strain “M2RS3-2” ) has been mounted and deposited at the Bayerische Staatssammlung für Paläontologie und Geologie in München , Germany , under voucher number SNSB-BSPG . GW30216 . Clonal individuals have been stored in ethanol as paratypes under voucher number SNSB-BSPG . GW30217 in addition to a DNA extraction under voucher number SNSB-BSPG . GW30218 . Etymology: hongkongensis , from “Hong Kong , ” and “-ensis , ” Latin , suffix referring to place of origin , as specimens are at present endemic to Hong Kong . The full name “Hoilungia hongkongensis” thus means “Hong Kong sea dragon . ” Two strains were used for this project: The “M2RS3-2” strain was used for the DNA sequencing ( the “DNA strain” ) and the “M153E-2” strain ( the “RNA strain” ) for the transcriptome . Both strains descend from a single placozoan individual each , which was isolated from mangroves/mangrove associates at two different sites in Hong Kong ( SAR , China ) . The DNA strain was isolated from a dead mussel shell collected in the Ho Chung River close to a small mangrove at Heung Chung village ( 22 . 352728N 114 . 251733E ) on June 6 , 2012 . The habitat undergoes daily changes in salinity , and on the day of collection , the salinity was 20 psu . The RNA strain was isolated from collection traps ( for details on slide sampling , see [82] ) connected to mangrove associates ( Hibiscus sp . ) and high shore mangrove ( Excoecaria sp . ) trees at Tai Tam Tuk ( 22 . 244708N 114 . 221978E ) on March 30 , 2012 . Both clonal cultures were cultured in 14 cm glass Petri dishes as described [19] , with a pure Pyrenomonas helgolandii algae culture ( strain ID 28 . 87 , Culture Collection of Algae , Georg-August-Universität Göttingen ) . The two different strains were used for DNA and RNA sequencing , respectively , to identify polymorphisms in these strains living in the same habitat but at two hydrogeographically distinct sampling sites ( northeast versus southeast Hong Kong ) . Animals were transferred in 20% BSA in artificial seawater , high-pressure frozen in a Wohlwend HPF Compact 02 , and stored in liquid nitrogen . Samples were processed from −90 °C to room temperature for Epon embedding in a Leica AFS unit as follows: they were fixed and contrasted in 0 . 1% tannic acid in acetone for 24 h and washed 4 times for 15 min in acetone; samples were then incubated in 2% Osmium tetroxide in acetone while the temperature was increased stepwise to −40 °C within the next 23 h; samples were then washed and progressively infiltrated in Epon:acetone mixes ( 1:2 , 2:1 ) and pure Epon while temperature was further raised from −40 °C to room temperature over 6 h . They were then polymerized in Epon . Seventy-nm ultrathin sections were cut on a Leica Ultracut and picked up on a copper slot grid 2 × 1 mm coated with a polystyrene film . Sections were poststained with uranyl acetate 2% in distilled water for 10 min , rinsed several times with distilled water followed by Reynolds lead citrate in distilled water for 10 min , and rinsed several times with distilled water . Micrographs were taken with a Transmission Electron Microscope Philips CM100 at an acceleration voltage of 80 kV with a TVIPS TemCam-F416 digital camera . A "lavalamp" kmer/GC plot was generated ( S2 Fig ) to yield a high-resolution plot of read counts per %GC and 31 bp kmer coverage using the Jellyfish kmer counter and a set of custom Python scripts ( kmersorter . py and fastqdumps2histo . py; for details on the procedure , see https://github . com/wrf/lavaLampPlot ) . In contrast to the conceptually similar approach Blobtools [108] , we used raw reads instead of contigs to yield a high-resolution plot of read counts per %GC and 31 mer coverage . The plot identified two read clouds with high counts at a kmer coverage of 80–140x ( heterozygous “read cloud” ) and 160–260x ( homozygous “read cloud” ) , respectively . Additional “read clouds” at 270–320x and 380–410x coverage mark repetitive sequence stretches . Another “read cloud” was found at a low coverage of 20–50x . Reads within this cloud and their pairs were extracted with kmersorter . py [added options: -s 0 . 16 -b 50 -w 0 . 40 -T -k 31] and fastqdumps2histo . py . Bowtie2 v2 . 2 . 5 [109] [added options: -q—no-sq] was used to map the 580 , 092 extracted reads to the 19 previously identified bacterial contigs ( see section “Contamination screening” ) . More than 86% of these reads mapped to the bacterial contigs , confirming the bacterial origin of the reads within the low-coverage “read cloud . ” Read counts identified a relatively high abundance of bacterial cells , and the GC content was similar to the host genome . To estimate the per-base genome coverage , paired-end reads were mapped to the softmasked reference assembly with Bowtie2 v2 . 2 . 5 [added options: -q—no-unal—no-sq ) and sorted with SAMtools v1 . 3 . 1 [110] . The bam file was used to create a bedgraph file in BEDtools v2 . 25 . 0 [111] by invoking the genomecov operation [added options: -ibam stdin -bga] . A custom Python script ( bedgraph2histo . py ) [added options: -m 2000] was used to create a coverage histogram table . Since 81 . 4% of the genome falls within the second peak ( 165–332x coverage with a maximum at 248x ) , most of the genome was merged in the reference assembly ( S3 Fig ) . To identify collinearity between the two placozoan species , all H . hongkongensis contigs >100 kb were aligned to the longest 10 T . adhaerens scaffolds ( accounting for 70 . 3 Mb or 66 . 5% of the genome assembly; including 5 . 7-Mb gaps ) with default settings . For generating the alignments , LASTZ v1 . 02 . 00 [115] ( implemented as a plugin in Geneious ) was used . Of the 222 H . hongkongensis contigs >100 kb , a total of 144 ( accounting for 60 . 6 Mb or 69 . 4% of the genome assembly ) aligned to the 10 longest T . adhaerens scaffolds . Aligned H . hongkongensis contigs were extracted from the assembly , sorted , and occasionally reverse complemented to be oriented according to the T . adhaerens scaffolds . Gene annotations ( GFF ) of contigs as well as protein sequences were extracted for the target scaffolds/contigs sets of both species . A MCScanX run [116] [added option: -a] was performed for each target set , using the extracted T . adhaerens and H . hongkongensis GFFs together with the reciprocal best 5 BLASTP hits [added options: -evalue 1e-10 -max_target_seqs 5 -outfmt 6] between and among proteins of both placozoans . Dual synteny line plots of the resulting collinearity files were visualized in VGSC v1 . 1 [117] [added options: -tp DualSynteny] and combined to Fig 2A . In addition , bar plots were generated for the 10 T . adhaerens scaffolds and the matching 144 H . hongkongensis contigs in VGSC [added option: -tp Bar] . Bar plots were mapped onto the DualSyntheny plots to show collinearity within each set and macrosynteny between both genomes . The percentage of collinearity between the T . adhaerens scaffolds and H . hongkongensis contigs was calculated in MCScanX , and results for the 10 scaffolds are given in S5 Table . The mean collinearity was calculated as the sum of the individual collinearities for the 10 T . adhaerens scaffolds multiplied by a size-correction faction for each scaffold ( i . e . , percent coverage of the evaluated 70 . 4 Mb of the T . adhaerens genome ) . Syntenic block sizes and the number of blocks were calculated using the custom Python script microsynteny . py ( described in [118] ) with skipping no more than 1 gene [added option: -s 1] and otherwise default options . To identify allele sharing or reproductive isolation , 3 genes encoding for ribosomal proteins were amplified via PCR , using degenerate primers designed based on the T . adhaerens genomic sequence , as well as a previously sequenced EST library of lineage H4 [19] . Primer sequences to amplify gDNA ( including intronic sequence ) for the ribosomal proteins L9 and L32 , as well as ribosomal protein P1 , were as follows: PCRs were run with an initial denaturation of 3 min at 94 °C; followed by 40 cycles of 30 s of denaturation at 94 °C , 30 s of annealing at 60 °C , and 1 . 5 min of elongation at 72 °C; and finished with a final elongation for 3 min at 72 °C . The BIOTAQ system was used ( Bioline , London ) . A list of samples used for amplification is provided as S6 Table . Sequencing was performed by Macrogen ( South Korea ) . Alleles were identified as double peaks in standard sequencing in the case of heterozygous alleles . The phasing of SNPs was inferred from homozygous sequences as well as the sequence of allelic variants in closely related haplotypes , for which phasing information was available because of the long Sanger reads . To check for reproductive isolation and to identify conspecific isolates , haplowebs [123] were generated for each marker as well as a CM [54] for combined markers using the online tool HaplowebMaker ( https://eeg-ebe . github . io/HaplowebMaker/; Spöri & Flot , in prep . ) . The resulting conspecificity scores were plotted in R using the heatmap3 package [124] , sorted according to a UPGMA tree ( JC69 model ) of the three concatenated genomic sequences ( with indels removed ) . If present , both alleles of an isolate were merged , and the consensus sequence was used to generate the tree . dN/dS ratios—as well as fractions of unchanged codons , synonymous , and nonsynonymous sites—were calculated based on a custom Python script ( alignmentdnds . py ) using regapped CDS alignments and untrimmed protein alignments ( S6 Fig ) . Codons with any ambiguous bases and gapped sites were ignored . Clustering into homologs and co-orthologs was performed with a custom python script ( makehomologs . py ) [added options:-s 1 -p 234 -H 200] . The script calls the MCL v12-068 algorithm [127] , which uses the output of a local all-versus-all BLASTP search [added options: -evalue 1e-3 -outfmt 6] of all H . hongkongensis and T . adhaerens proteins . To identify enriched GO terms in non-BLAST hits as well as in four co-ortholog categories ( one-to-many , many-to-one , many-to-many , and many-to-zero ) , an enrichment analysis was performed for the three main GO categories ( Biological Process , Cellular Component , and Molecular Function ) using topGO [128] . Only enriched GO terms with a p-value <0 . 05 were kept , based on the classic Fisher test . Ortholog categories ( see also [40] ) are defined as ( 1 ) one-to-one: Only one ortholog is found in each species; ( 2 ) one-to-many: One ortholog in this species , but many co-orthologs in the other species . The gene was duplicated in the other species from the ancestral copy after speciation; ( 3 ) many-to-one: More than one co-ortholog in this species but only one in the other species . The gene was duplicated in this species from the ancestral copy after speciation; ( 4 ) many-to-many: More than one co-ortholog in this and the other species . At least two gene duplications could be found from an ancestral gene in the common ancestor of both species—one duplication in this species , and a second one in the other species; ( 5 ) many-to-zero: Many co-orthologs in this species but none in the other . In this case , the gene was duplicated from an ancestral copy in this species after speciation and likely lost in the other species . To estimate molecular differences between H . hongkongensis and T . adhaerens and to bring these into a taxonomic context , we measured genetic distance using an extended data matrix of 212 nuclear proteins set up by Cannon and colleagues [55] . This data matrix was chosen as it includes a comparable number of sites for a diverse taxonomic range and is , therefore , also suitable for phylogenetic analyses . In addition , genetic distances were measured for 5 standard barcoding ( “selected” ) markers—namely , nuclear ribosomal subunits 18S ( S9 Fig ) and 28S ( S10 Fig ) , mitochondrial large ribosomal subunit 16S ( S11 Fig ) , and the mitochondrial proteins cytochrome c oxidase subunit 1 ( CO1 ) ( S12 Fig ) and NADH dehydrogenase subunit 1 ( ND1 ) ( S13 Fig ) . An overview of means for all distances of all six marker sets is provided as S14 Fig . The incorporation of datasets from four individual categories ( nuclear protein versus nuclear rDNA versus mitochondrial protein versus mitochondrial rDNA ) enabled the comparison among markers with different substitution rates . To assess the effect of adding a second placozoan species on the placement of the Placozoa in the animal tree of life and to estimate branch lengths to the two placozoan species , dataset 1 was further condensed to generate a highly complete protein matrix ( dataset 2 ) . This set had only 10 . 8% missing characters in 58 taxa , including 32 nonbilaterians and 2 outgroups with an almost complete gene set ( 194 genes , see also gene density matrix in S21 Fig ) . It has been demonstrated that the CAT model ( specifically CAT-GTR ) implemented in PhyloBayes [132] fits phylogenomic amino acid supermatrices containing nonbilaterians best [73 , 133] , and obviously , only best-fitting evolutionary models should be used in probabilistic phylogenetic analyses to reduce systematic errors [133] . However , the computational burden of reaching convergence of analyses using the CAT-GTR model can be prohibitive . It is also well known that phylogenomic datasets frequently suffer from compositional heterogeneity that might negatively influence phylogeny estimation [134–136] . Compositional heterogeneities can be reduced by the so-called Dayhoff recoding [65 , 137 , 138] , which combines amino acids with similar physicochemical properties into one of six categories . Through this reduction of character space , lineage-specific compositional heterogeneities are lessened—at the cost , however , of losing phylogenetic signal [67] . However , another advantage of Dayhoff recoding is a significant reduction of computation time needed to reach convergence . The protein as well as the Dayhoff 6-state recoded dataset 2 were analyzed with PhyloBayes-MPI v1 . 7 [69 , 132] , employing the CAT-GTR model , on the Linux cluster of the Leibniz Rechenzentrum ( http://www . lrz . de ) in Garching bei München , running 2 chains ( each on 112 CPUs ) each until reaching convergence , as estimated by using tracecomp and bpcomp programs of the PhyloBayes package ( see PhyloBayes manual for details ) . Furthermore , to evaluate the effect of using less-fitting site-homogeneous evolutionary models on the phylogenetic relationships of the Placozoa , we conducted a PhyloBayes-MPI analysis as above but with the GTR model ( see for example [73] , [68] ) , and also two maximum-likelihood analyses in RAxML: one with the GTR model using RAxML-NG v0 . 5 . 1b [139] [added options:—model PROTGTR+G+I—bs-trees 100—data-type AA] and one with the LG model using RAxML v8 . 2 [70] [added options: -f a -x 670 -m PROTGAMMAILG -p 220 -N 100] . The LG model was used as it was the best-fitting site-homogeneous model in 210 of the 212 gene partitions determined by ProtTest v3 . 4 [140] . Phylogenetic trees are shown as S15–S20 Figs . | Placozoans are a phylum of tiny ( approximately 1 mm ) marine animals that are found worldwide in temperate and tropical waters . They are characterized by morphological simplicity , with only a handful of cell types , no neurons , no tissue organization , and even no axial polarity . Since the original description of Trichoplax adhaerens 135 years ago , no additional accepted species has been established , leaving the Placozoa as the only animal phylum with only a single formally described species . While classical morphological species identification has failed to reveal further species , single-gene DNA sequence analyses have identified a broad and deep genetic diversity within the Placozoa . To address the significance of this deep genetic diversity in this morphologically uniform phylum , and to better understand its consequences for speciation processes , general biology , and species delimitation in the Placozoa , we sequenced the genome of the placozoan isolate “H13 , ” a lineage distantly genetically related to T . adhaerens . Our multilevel genomic comparisons with the T . adhaerens genome show considerable differences in the general structure of the genome and the makeup and history of various gene families of biological relevance to habitat adaptation . Based on comparative genomics , we here describe the second placozoan species and show that it belongs to a new genus . | [
"Abstract",
"Introduction",
"Results",
"and",
"discussion",
"Material",
"and",
"methods"
] | [
"taxonomy",
"invertebrates",
"animals",
"trichoplax",
"adhaerens",
"placozoa",
"phylogenetics",
"data",
"management",
"phylogenetic",
"analysis",
"energy-producing",
"organelles",
"mitochondria",
"bioenergetics",
"cellular",
"structures",
"and",
"organelles",
"research",
"and",
"analysis",
"methods",
"sequence",
"analysis",
"computer",
"and",
"information",
"sciences",
"sequence",
"alignment",
"bioinformatics",
"comparative",
"genomics",
"short",
"reports",
"evolutionary",
"systematics",
"biochemistry",
"eukaryota",
"cell",
"biology",
"cnidaria",
"database",
"and",
"informatics",
"methods",
"genetics",
"biology",
"and",
"life",
"sciences",
"genomics",
"evolutionary",
"biology",
"computational",
"biology",
"organisms"
] | 2018 | Comparative genomics and the nature of placozoan species |
Filamentous fungi rely heavily on the secretory pathway , both for the delivery of cell wall components to the hyphal tip and the production and secretion of extracellular hydrolytic enzymes needed to support growth on polymeric substrates . Increased demand on the secretory system exerts stress on the endoplasmic reticulum ( ER ) , which is countered by the activation of a coordinated stress response pathway termed the unfolded protein response ( UPR ) . To determine the contribution of the UPR to the growth and virulence of the filamentous fungal pathogen Aspergillus fumigatus , we disrupted the hacA gene , encoding the major transcriptional regulator of the UPR . The ΔhacA mutant was unable to activate the UPR in response to ER stress and was hypersensitive to agents that disrupt ER homeostasis or the cell wall . Failure to induce the UPR did not affect radial growth on rich medium at 37°C , but cell wall integrity was disrupted at 45°C , resulting in a dramatic loss in viability . The ΔhacA mutant displayed a reduced capacity for protease secretion and was growth-impaired when challenged to assimilate nutrients from complex substrates . In addition , the ΔhacA mutant exhibited increased susceptibility to current antifungal agents that disrupt the membrane or cell wall and had attenuated virulence in multiple mouse models of invasive aspergillosis . These results demonstrate the importance of ER homeostasis to the growth and virulence of A . fumigatus and suggest that targeting the UPR , either alone or in combination with other antifungal drugs , would be an effective antifungal strategy .
Aspergillus fumigatus is a soil-dwelling filamentous fungus that has become the predominant mold pathogen of the immunocompromised population [1] , [2] , [3] . The infection is acquired through the inhalation of aerosolized conidia ( spores ) , which are small enough to reach the distal airways [4] . When the inhaled conidia germinate and develop into hyphae , secreted fungal hydrolases progressively damage the integrity of the pulmonary epithelium , allowing vascular invasion with subsequent hematogenous spread [4] , [5] . Despite the introduction and use of recently approved antifungals , invasive aspergillosis ( IA ) continues to be associated with a poor outcome [1] , [3] , [6] , [7] . Moreover , the incidence of IA is expected to rise with the expansion of the immunosuppressed population , making the search for novel treatments a high priority . Unfortunately , few effective drugs are identifiable in the late-stage development pipeline [8] , emphasizing the need for increased understanding of the virulence of this organism to facilitate the rational design of novel therapeutic strategies . The prevailing evidence suggests that the virulence of A . fumigatus involves gene products that have evolved to enhance the competitiveness of the fungus in the ecologically diverse environmental niche of decaying organic debris . The saprophytic nature of this lifestyle requires the secretion of abundant enzymes that enable the fungus to extract nutrients from complex polymeric material [5] , [9] , [10] . This high capacity secretory system has been exploited in other filamentous fungi for the industrial production of native and heterologous proteins and is a feature that distinguishes these organisms from the yeast Saccharomyces cerevisiae [11] , [12] , [13] , in which secretion levels are sometimes too low for industrial application [14] . As in all eukaryotes , the endoplasmic reticulum ( ER ) of filamentous fungi is the major processing center for secreted and transmembrane proteins . The unique environment of the ER facilitates the folding of nascent proteins , a process that is aided by ER-resident chaperones and folding enzymes , and post-translational modifications such as glycosylation , phosphorylation and disulfide bridge formation [15] . From the ER , proteins are transferred to the Golgi compartment , where they undergo further processing before being delivered to the membrane by vesicles of the distal secretory system . Under normal conditions , the secretory demand is balanced by the protein folding capacity of the ER . However , as much as a third of newly synthesized proteins fail to achieve native structure due to imperfections in transcription , translation , post-translational modifications or protein folding [16] . Thus , an inevitable consequence of a high rate of protein synthesis , coupled with a rapid flux through the secretory system , is the accumulation of misfolded proteins in the ER . Unfolded proteins threaten cell survival because they form toxic aggregates that interfere with the function of normal proteins [17] . If the influx of nascent unfolded polypeptides exceeds the folding capacity of the ER , the ensuing ER stress triggers a series of adaptive responses collectively termed the unfolded protein response ( UPR ) [18] . The UPR is a conserved eukaryotic signaling pathway that originates in the ER and transmits information on the folding capacity of the secretory system to the nucleus . Upon activation , the UPR restores ER homeostasis by reducing the flow of proteins into the ER , increasing protein transport out of the ER , increasing the expression of ER-resident chaperones and foldases , and by degrading proteins that fail to properly fold [18] , [19] . The UPR affects the secretory pathway at multiple levels and has been shown to involve at least 381 genes in S . cerevisiae , underscoring the complexity and importance of this pathway for cell survival [20] . The upstream ER sensor responsible for detecting unfolded proteins and triggering the UPR is Ire1p , a transmembrane protein that has an ER lumenal sensing domain and a protein kinase and endoribonuclease domain in the cytoplasmic region [21] . Activation of the sensor is triggered by interaction with unfolded proteins in the ER lumen , possibly facilitated by the dissociation of ER-resident chaperones from Ire1p [19] . These events elicit Ire1p aggregation and trans-autophosphorylation , resulting in activation of the cytosolic ribonuclease domain [22] , [23] . The increased ribonuclease activity catalyzes the spliceosome-independent cleavage of the cytoplasmic precursor mRNA HAC1u ( uninduced ) , removing a single intron to generate the induced form of the HAC1 mRNA , HAC1i ( induced ) [24] . This unconventional splicing event creates a frame-shift in the mRNA , allowing for the translation of a transcription factor that moves to the nucleus and regulates the expression of UPR target genes [20] , [25] . Despite advances in our understanding of the nature and scope of the UPR in several organisms , the significance of this pathway to the virulence of a pathogenic eukaryote is unknown . In this study , we generated a mutant of A . fumigatus that is deficient in UPR signaling by disrupting the hacA gene , encoding the ortholog of the yeast Hac1p transcription factor responsible for modulating the expression of UPR target genes . The results demonstrate that A . fumigatus relies heavily on the UPR to sustain growth under conditions that disrupt ER homeostasis , including thermal stress , cell wall stress and a high secretory load . In addition , loss of the UPR was associated with attenuated virulence and a dramatic increase in antifungal drug sensitivity . Taken together , these data provide evidence that A . fumigatus is under ER stress in vivo and would , therefore , be vulnerable to therapeutic attack on the fungal UPR .
The UPR-induced ( hacAi ) and uninduced ( hacAu ) forms of the A . fumigatus hacA mRNA were cloned by RT-PCR from RNA derived from cultures grown in the presence or absence of dithiothreitol ( DTT ) -induced ER stress , respectively . A comparison of the cDNA sequences with the A . fumigatus genome revealed a conventional intron with consensus border sequences that is excised in both the hacAu and hacAi mRNAs . In addition , an unconventional 20 nt intron is uniquely excised from the A . fumigatus hacAi mRNA in response to ER stress ( Figure 1A ) , similarly to what has been reported in other orthologs of this mRNA [26] . This atypical intron is much smaller than the corresponding 252 nt intron of S . cerevisiae HAC1 [24] , but is similar in size to introns that are spliced by UPR activation in mammals ( 26 nt ) [27] , Caenorhabditis elegans ( 23 nt ) [27] , Candida albicans ( 19 nt ) [28] and filamentous fungi ( 20 nt ) [29] , [30] . The exact splicing sites of the unconventional intron in A . fumigatus could not be unambiguously identified by comparing cDNA and genomic sequences because of the presence of a CTGCAG at each side of the intron , a feature that is also found in other filamentous fungi [29] , [30] . Although the size of the intron varies between genera , the border sequences are highly conserved ( Figure 1A ) and are located in a region of strong predicted RNA secondary structure ( data not shown ) [29] , [30] . The first 213 amino acids encoded by the hacAu and hacAi mRNAs are identical . This region contains a leucine zipper dimerization motif adjacent to a basic DNA binding domain ( Figure 1B ) , which is characteristic of bZIP-type family transcription factors . The atypical splicing of the A . fumigatus hacAi mRNA changes the reading frame , resulting in an encoded protein that replaces 220 amino acids at the c-terminus of HacAu with a unique c-terminal domain comprised of 129 amino acids ( Figure 1B ) . The resulting HacAi protein has 76% and 81% identity to the corresponding proteins in A . nidulans and A . niger , respectively . Alignment of the A . fumigatus HacAi protein with orthologs from other filamentous fungi reveals extensive homology throughout the protein ( Figure 1C ) . By contrast , most of the homology to S . cerevisiae Hac1p is concentrated in the DNA binding domain ( data not shown ) . Deletion of hacA was accomplished by replacing the hacAi open reading frame with the hygromycin resistance cassette ( Figure 2 ) . To determine whether loss of hacA was sufficient to disrupt UPR signaling in A . fumigatus , the expression of four known UPR target genes was examined by northern blot analysis , including bipA ( ER chaperone ) , pdiA and tigA ( protein disulfide isomerases ) and hacA itself . Each of these genes contains an unfolded protein response element ( UPRE ) in its promoter [31] , and the abundance of each mRNA increases in response to UPR activation [30] . As expected , treatment of wt A . fumigatus with DTT increased hacA abundance and induced the conversion of hacAu into hacAi , indicating activation of the UPR under these conditions ( Figure 3A ) . The smaller size of the hacAi mRNA is consistent with a 5′ mRNA truncation that has been reported following UPR induction in other filamentous fungi [29] , [30] . In contrast to wt A . fumigatus , the ΔhacA mutant was unable to increase the level of three other UPR target genes when treated with DTT , indicating a defect in UPR-regulated gene expression . Complementation of the ΔhacA mutant ( C' ) restored UPR signaling to the ΔhacA mutant ( Figure 3B ) . To determine how loss of hacA impacts the growth of A . fumigatus under conditions of ER stress , the mutant was incubated in the presence of agents that disrupt ER homeostasis by different mechanisms , including DTT , tunicamycin ( TM ) , and brefeldin A ( BFA ) . DTT unfolds proteins directly by reducing disulfide bonds , TM impairs protein folding by inhibiting N-linked glycosylation , and BFA impairs anterograde protein transport from the ER to the Golgi [26] . The growth of the ΔhacA mutant was comparable to wt in the absence of ER stress , although conidiation was decreased on solid medium ( Figure 4 , columns marked ‘0' ) . However , the ΔhacA mutant was unable to grow in the presence of concentrations of DTT , BFA or TM that could be tolerated by wt A . fumigatus , indicating heightened sensitivity to ER stress . The mutant was also hypersensitive to the superoxide-generating agent paraquat ( Figure S1 ) , consistent with the adverse effects of oxidative stress on protein folding and ER homeostasis [32] . The radial growth rate of the ΔhacA mutant was almost indistinguishable from that of wt A . fumigatus at 37°C or 42°C ( Figure 5A and Figure S2 ) . However , at 45°C the ΔhacA mutant failed to grow beyond the site of the initial inoculum ( Figure 5B ) . Microscopic analysis revealed that the ΔhacA conidia had germinated at 45°C , but subsequently arrested growth as young hyphae ( data not shown ) . To determine whether this was due to loss of viability , 200 conidia were evenly distributed onto an agar surface . After incubating at 45°C for 0 , 12 and 24 h , the plates were shifted to 37°C , and surviving colony forming units ( CFUs ) were counted . As shown in Figure 5C , approximately 50% of the plated wt and complemented conidia survived 24 h of incubation at 45°C . This was in contrast to the ΔhacA mutant , where less than 1% of the plated conidia survived 24 h at 45°C . A . fumigatus is a thermotolerant fungus that normally thrives at temperatures above 50°C [33] , with an optimum for growth between 37°C and 42°C , depending on the medium . The ΔhacA mutant was unable to grow at a temperature that is only 3°C above the optimum range for this species . This phenotype was unexpected because temperature-induced lethality has not been previously reported in the corresponding Δhac1 mutant of S . cerevisiae [20] , [34] , suggesting a vulnerability in A . fumigatus that may not be present in yeast . S . cerevisiae grows optimally between 25°C and 30°C , and 37°C is considered thermal stress for this organism [35] . We found that the growth of a S . cerevisiae Δhac1 mutant was indistinguishable from that of wt at either 30°C or 37°C , indicating that yeast differ from A . fumigatus in their ability to tolerate thermal stress in the absence of proper UPR signaling ( Figure 5D ) . The inability of the ΔhacA mutant to grow at 45°C could be reversed by osmotic stabilization of the medium with sorbitol ( Figure 6A ) or KCl ( data not shown ) , suggesting that the impaired growth of ΔhacA at elevated temperature is due , in part , to loss of cell wall integrity . To test this more directly , conidia were inoculated onto cover-slips in liquid medium and germinated overnight at 37°C . After shifting to 45°C , the hyphae were examined microscopically . Within 4 h of incubation at the elevated temperature , the hyphal tips began to swell , and tip lysis became apparent within 8 h ( Figure 6B ) . Occasional areas of cytoplasmic leakage were also observed in subapical hyphae , possibly representing emerging branch points ( data not shown ) . The ΔhacA mutant was severely growth impaired following the temperature shift , suggesting an important role for HacA in the maintenance of cell wall integrity at the hyphal tips during thermal stress . Since thermal stress is likely to have pleiotropic effects on cell physiology , calcofluor white ( CFW ) was used as a more specific inhibitor of cell wall integrity . CFW is an anionic dye that weakens the wall by binding to nascent chitin chains [36] . Northern blot analysis revealed that treatment with CFW induces the hacA-dependent accumulation of bipA mRNA ( Figure 7A ) , suggesting that the UPR is part of the normal adaptive response to CFW-induced cell wall stress . The ΔhacA mutant was unable to grow in the presence of concentrations of CFW that had minimal effect on the wt or complemented strains , consistent with a protective role for HacA under these conditions ( Figure 7A ) . Normal growth could be restored to the mutant by osmotic stabilization of the medium with sorbitol , supporting the notion that the impaired growth of the mutant in the presence of CFW was a consequence of reduced cell wall integrity ( Figure 7B ) . Microscopic analysis of the ΔhacA mutant in the presence of CFW revealed the same apical lysis that was observed under conditions of thermal stress ( data not shown ) , suggesting that the mutant wall is particularly vulnerable to cell wall perturbation at the tips . Similar results were obtained using the related cell wall damaging compound Congo red ( Figure S3 ) . However , these findings in A . fumigatus were in contrast to S . cerevisiae , where the corresponding Δhac1 mutant showed wt sensitivity to CFW ( Figure 7C ) . The increased vulnerability of the ΔhacA mutant to cell wall stress raises the possibility that HacA contributes to cell wall homeostasis in A . fumigatus . To test this , a biochemical analysis of the cell wall was performed . As shown in Table 1 , the ΔhacA mutant revealed a significant decrease in glucose content in both the alkali insoluble ( AI ) and alkali soluble ( AS ) fractions of the cell wall relative to wt , suggesting a defect in both β ( 1–3 ) and α ( 1–3 ) glucan composition in the mutant cell wall . All major classes of antifungal drugs that are currently in use against A . fumigatus attack the integrity of the membrane or cell wall . Fungi respond to these agents by upregulating cell wall and membrane repair systems [37] , [38] , [39] , which may increase stress on the secretory system . To determine how loss of UPR function would affect growth in the presence of antifungal stress , susceptibility to amphotericin B , caspofungin , itraconazole and fluconazole was compared using the Etest method . Conidia were spread onto the surface of an agar plate , and 4 Etest strips were placed on top , each impregnated with a concentration gradient of a different antifungal drug , and incubated at 37°C for 48 h . The ΔhacA mutant had larger areas of growth inhibition surrounding each strip , indicating heightened susceptibility to each of these drugs and a decrease in the minimal inhibitory concentration ( Figure 8A ) . The incomplete clearing around the caspofungin strip on plates inoculated with the wt or complemented strains is consistent with the known fungistatic activity of this class of drug for A . fumigatus [40] . It is therefore striking that a complete zone of clearing was evident around the caspofungin strip on the plate inoculated with the ΔhacA mutant . Agar plugs taken from the zone of growth inhibition surrounding the caspofungin strip on wt-inoculated plates were able to grow when transferred to medium lacking any drug . However , no viable organism could be recovered from agar plugs taken from the cleared zone surrounding the caspofungin strip on ΔhacA-inoculated plates , indicating that caspofungin becomes fungicidal in the absence of UPR function . Microscopic analysis of caspofungin-treated hyphae revealed normal morphology in the wt , but abnormal swelling and lysis in the ΔhacA mutant ( Figure 8B ) . These defects were localized to hyphal tips and branch points , similar to what was observed under conditions of thermal stress and CFW treatment . This experiment was performed on RPMI agar in accordance with the manufacturer's specifications , but comparable results were also obtained using IMA as the medium ( Figure S4 ) . Remarkably , the corresponding Δhac1 mutant in S . cerevisiae did not show increased sensitivity to either caspofungin , ketoconazole , amphotericin B or fluconazole ( Figure S5 ) . The ability of A . fumigatus to colonize the host begins with the germination of conidia in the lung followed by invasion of exploring hyphae into the surrounding tissue . The organism must acquire nutrients from host tissues at all steps of the infection , which requires continual secretion of a multitude of degradative enzymes . Since ER stress occurs when protein secretion is upregulated [41] , we hypothesized that loss of UPR signaling would impair the secretory capacity of A . fumigatus . To test this prediction , secreted proteolytic activity was quantified with the Azocoll assay , using conditions previously described for A . fumigatus [10] . Azocoll is an insoluble collagen linked to an azo dye , and its hydrolysis releases soluble colored peptides that can be quantified colorimetrically [42] . As shown in Figure 9A , culture supernatants derived from the ΔhacA mutant were significantly less efficient at digesting Azocoll than wt cultures , indicating that protease secretion is abnormal in the mutant . This decrease in proteolytic activity was consistent with an overall reduction in secreted protein levels in the ΔhacA mutant , as revealed by SDS-PAGE analysis of culture supernatants ( Figure 9A ) . The reduced secretory capacity of the ΔhacA mutant predicts that this strain would have difficulty assimilating nutrients from a complex substrate . On IMA medium , the growth of the ΔhacA mutant was normal ( Figure 5A ) . However , this rich medium contains a substantial amount of reduced carbon and nitrogen in the form of tryptone ( pancreatic digest of casein ) , peptone ( enzymatic digest of proteins ) , yeast extract , dextrose and starch . When challenged to use a more complex substrate such as skim milk , the ΔhacA mutant grew slower than the wt and complemented strains ( Figure 9B ) . Osmotic stabilization with sorbitol was unable to rescue this phenotype , but the addition of a reduced nitrogen/carbon source completely restored growth to wt levels ( Figure 9B , and data not shown ) . These findings argue that the impaired growth of ΔhacA on skim milk agar is due to inefficient nutrient acquisition rather than an indirect effect on cell wall stress . Similar observations were made when mouse lung tissue was used as a substrate . In contrast to the wt-inoculated lung tissue , which supported fungal growth within 24 h , the ΔhacA-inoculated lung showed no signs of fungal growth ( Figure 9C ) . Collectively , these findings suggest that the UPR promotes the growth of A . fumigatus on complex polymeric material by facilitating the production of secreted hydrolases that are necessary to breakdown the substrate into usable nutrients . The ability to detect A . fumigatus proteases in vivo [43] implies that active secretion occurs in the host environment , suggesting that the UPR may contribute to virulence . To test this , we compared the virulence of the ΔhacA mutant to that of wt A . fumigatus . As shown in Figure 10A , the ΔhacA mutant was hypovirulent in an outbred mouse model of invasive aspergillosis that uses a single dose of triamcinolone acetonide ( TA ) to induce a period of transient immunosuppression . An increasing body of evidence suggests that the outcome of virulence testing in experimental models of aspergillosis is influenced by host strain and the type of immunosuppression [44] , [45] , [46] . Thus , virulence was also compared in two additional models that use inbred mice: a neutropenic model and a cortisone acetate model . The ΔhacA mutant had attenuated virulence in all three model systems , demonstrating that the UPR is an important stress signaling pathway in the host environment ( Figure 10A , 10B , and 10C ) .
All eukaryotes with an elevated capacity for protein production depend on the UPR to maintain ER homeostasis . ER stress is encountered under many adverse environmental conditions that cause protein unfolding , including high temperature [47] , [48] , oxidative stress [49] , [50] , hypoxia [51] , [52] or nutrient limitation [53] . However , it may also occur under normal physiological conditions in response to a change in the demand for secretion . For example , the UPR is induced in B cells when they are stimulated to secrete antibody [54] . Similarly , UPR-deficient mice fail to differentiate hepatocytes , pancreatic β cells or plasma cells , because the UPR protects against the ER stress that is generated when the intense secretory activity of these cells is activated [55] , [56] , [57] . A . fumigatus , like many other filamentous fungi , is well equipped for protein secretion [5] , [9] , [58] , with over 1% of its genome dedicated to secreted proteases alone [59] , [60] . This robust secretory arsenal makes filamentous fungi excellent production hosts for proteins of biotechnology interest [11] , [12] , [13] , [61] , and the ability of enforced HacAi overexpression to further enhance protein secretion [62] illustrates the importance of the UPR to the maintenance of ER homeostasis under a high secretory load . In this study , a UPR-deficient mutant of A . fumigatus was constructed in order to determine how the UPR impacts growth , secretion and virulence in A . fumigatus . The ΔhacA mutant was hypersensitive to agents that perturb ER homeostasis and was unable to increase the expression of four known UPR target genes in response to ER stress . Although a hacA-independent mechanism of bipA induction has been recently reported in A . niger strains that overproduce membrane proteins [63] , the absence of bipA induction in the ΔhacA mutant treated with DTT indicates that bipA induction is hacA-dependent when DTT is used to induce ER stress . In S . cerevisiae , Δhac1 mutants are inositol auxotrophs , a phenotype that is associated with defects in expression of INO1 [34] . However , inositol was dispensable for the growth of the A . fumigatus ΔhacA mutant ( data not shown ) , indicating that hacA is not required for this pathway in A . fumigatus . On rich medium , the growth of the ΔhacA mutant was comparable to that of wt . However , the ΔhacA mutant became growth impaired when it was forced to obtain nutrients from complex substrates such as skim milk or mouse lung tissue . These findings argue that a rapid growth rate per se does not constitute sufficient ER stress to require extensive support from the UPR , as long as an adequate supply of reduced nitrogen/carbon is present . By contrast , growth on more polymeric material would require an increase in secretory activity to breakdown the substrate , resulting in UPR activation . Communication from other stress response pathways may also influence the magnitude of this response , such as the ability of Gcn4p to control both amino acid starvation responses and UPR target gene expression [64] . Failure to trigger the UPR under situations that demand increased secretory capacity would be expected to impair secretion , thereby limiting nutrient availability and reducing growth . Furthermore , the unresolved ER stress caused by loss of UPR function may trigger a second feedback mechanism that is activated in response to impaired secretory protein folding or transport called repression under secretion stress ( RESS ) [65] . RESS involves the selective transcriptional downregulation of genes encoding certain secreted proteins , and current evidence suggests that it is controlled differently from the UPR [66] . Our finding of reduced secreted collagenolytic activity and an alteration in the overall secretory profile of the ΔhacA mutant is consistent with this model ( Figure 9A ) . In nature , A . fumigatus has evolved to thrive in compost , an environmental niche that generates heat from microbial activity . A . fumigatus has acquired unique mechanisms of thermotolerance to support its growth up to 60°C [33] . This study demonstrates that the UPR plays an essential role in thermotolerance . The ΔhacA mutant grew normally at 37°C but was unable to maintain cell wall integrity at 45°C . Cytoplasmic leakage was observed at hyphal tips and at various points along the hypha , possibly representing areas of weakness caused by dynamic remodeling of the cell wall at these sites . Reduced thermotolerance has also been reported in other cell wall mutants of filamentous fungi , suggesting that thermal stress has adverse effects on the cell wall [67] , [68] . Prolonged incubation of the ΔhacA mutant at 45°C was incompatible with viability . This finding is particularly notable in view of the extraordinary thermotolerance of A . fumigatus , but is also remarkable because the corresponding mutation in S . cerevisiae does not exhibit the same temperature-sensitive phenotype ( Figure 5 ) . This difference may reflect the need for more surface export functions in A . fumigatus to maintain the integrity of the apical cell wall during polarized growth , particularly at elevated temperature . Direct perturbation of the cell wall by treatment with CFW increased apical lysis and reduced hyphal growth in the ΔhacA mutant . Sorbitol rescued this CFW sensitivity , suggesting that the phenotype is primarily a consequence of reduced cell wall integrity ( Figure 7B ) . By contrast , neither the temperature sensitivity nor the reduced conidiation of the ΔhacA mutant could be fully rescued by sorbitol ( Figure 6A ) , suggesting that the UPR has additional homeostatic functions under conditions of thermal stress that do not involve the cell wall . The components of the A . fumigatus cell wall can be divided into two main groups based on their alkali solubility [69] . The AI fraction is thought to provide the main structural rigidity of the wall and is composed of β ( 1–3 ) glucan , chitin and galactomannan . By contrast , the AS fraction contains predominantly α ( 1–3 ) glucan and galactomannan . The heightened sensitivity of the ΔhacA mutant to multiple types of cell wall stress , combined with the decreased glucose content in its cell wall , is consistent with a defect that reduces the overall glucan composition of the cell wall . The A . fumigatus wall is a highly dynamic structure , particularly at hyphal tips and branch points where the structural needs of the hypha must be balanced by the demand for new apical growth [69] . Since β ( 1 , 3 ) glucan synthase is transported to the growing tips as an inactive complex through the secretory pathway [70] , it is intriguing to speculate that the predisposition of the ΔhacA mutant to apical lysis is due to inefficient delivery of the glucan synthase complex to the growing tips . The ΔhacA mutant had normal chitin levels however , suggesting that HacA is less important for chitin synthase activity . This may reflect the ability of multiple chitin synthases [71] to compensate for any reduction in chitin synthase delivery caused by loss of HacA . Interestingly , the S . cerevisiae Δhac1 mutant had wt sensitivity to CFW , suggesting a fundamental difference between these two species in terms of their reliance on the HacA-dependent UPR for cell wall homeostasis . Analysis of the antifungal susceptibility profile of the ΔhacA mutant revealed two important findings . First , the ΔhacA mutant showed a dramatic increase in susceptibility to antifungal drugs that are in use for the treatment of invasive aspergillosis . This suggests that targeting the UPR with novel therapy could act synergistically with currently approved antifungal drugs , as well as potentially increasing the susceptibility profile of other fungal pathogens that are intrinsically resistant to some antifungals . For example , A . terreus and most Fusarium and Scedosporium isolates are only moderately susceptible or resistant to amphotericin B [72] , [73] , [74] , [75] . Similarly , itraconazole has limited activity against Fusarium and Scedosporium species , and voriconazole and echinocandins are largely ineffective against zygomycetes [72] , [76] . Thus , targeting the UPR has the possibility of expanding the number of therapeutic options for these emerging fungal pathogens . The second important observation is that the well known fungistatic effects of the β ( 1–3 ) glucan synthase inhibitor caspofungin became fungicidal to A . fumigatus in the absence of hacA function . This synergistic activity is likely to reflect the lethal effects of glucan synthase inhibition in a strain that is already deficient in glucan production . These observations also suggest that the UPR is an essential component of the adaptive response to antifungal stress , an idea that is supported by the upregulation of genes involved in ER and secretion functions , including Hac1 , during caspofungin treatment of the dimporphic yeast Candida albicans [77] . This class of genes was not induced by caspofungin treatment of S . cerevisiae , suggesting an important difference between these yeasts [39] , [78] . In addition , loss of UPR function increases sensitivity to cell wall damage in A . fumigatus ( Figure 5 ) and in C . albicans [79] , but not in S . cerevisiae ( Figure 5 ) . One possible explanation for these differences is that the ability to form true hyphae in A . fumigatus and C . albicans increases the demand on the secretory system for cell wall repair , making these fungi more vulnerable to loss of UPR function . Since recently published data have shown that the virulence of A . fumigatus is influenced by host strain and the type of immunosuppression , three distinct mouse models of invasive aspergillosis were used to assess virulence . The ΔhacA mutant was hypovirulent in all three models , emphasizing the importance of UPR signaling to the ability of the fungus to grow in the host environment . Metabolic evidence has suggested that A . fumigatus relies heavily on protein degradation as a major source of nutrients in vivo [80] , which is consistent with the detection of secreted A . fumigatus proteases in vivo [43] . Although protease secretion has long been considered a virulence-related factor for A . fumigatus , single-gene disruptions have yet to demonstrate this because of the abundant secreted proteases encoded by the genome . Here , we provide the first genetic evidence to suggest that secretory activity is important to the virulence of this organism . The results are consistent with a model in which the UPR contributes to virulence by supporting the secretory activity that is necessary to degrade host tissues . In the absence of a functional UPR , this secretory capacity is impaired , which may lessen the ability of the organism to damage tissues and efficiently extract the nutrients required for growth . Failure to resolve ER stress could also contribute to the reduced virulence of the ΔhacA mutant if unfolded proteins accumulate to toxic levels . Taken together , the data from this study suggest that the high secretory capacity of A . fumigatus places it at considerable risk for ER stress and thus represents a vulnerability that could be exploited for therapeutic gain by disrupting the pathways that maintain ER homeostasis . Moreover , since secretory processes have prominent roles in the virulence of parasitic protozoa [81] , [82] , the findings from this study may have relevance to other pathogenic eukaryotes .
The A . fumigatus and S . cerevisiae strains used in the study are listed in Table 2 . Conidia were harvested from Aspergillus minimal medium ( AMM ) [83] containing 10 mM ammonium tartrate and osmotically stabilized with 1 . 2 M sorbitol . Unless otherwise specified , experiments involving the ΔhacA mutant were performed on inhibitory mold agar ( IMA , Fisher Scientific Cat . # 14-910-95 ) since the growth rate of the ΔhacA mutant approximated that of wt on this medium . Radial growth rate was determined by spotting 5 , 000 conidia onto the center of a plate and monitoring colony diameter daily . For analysis of survival under thermal stress , 200 conidia were evenly spread onto the surface of an IMA plate . After incubating at 45°C for 0 h , 12 h , or 24 h , the plates were placed at 37°C and surviving colony forming units ( CFUs ) were counted after 24 h of growth . To demonstrate cytoplasmic leakage at 45°C , conidia were inoculated onto a glass coverslip submersed in AMM and incubated at 37°C for 24 h . After shifting to 45°C for 4 h and 8 h , the coverslip was inverted onto a glass slide and the hyphae were photographed by differential interference contrast ( DIC ) microscopy . To monitor growth under ER or cell wall stress , 2 , 000 conidia were inoculated onto the center of a plate containing IMA supplemented with concentrations of BFA , TM , CFW , or Congo red specified in the Results section . The plates were incubated for 2–3 days at 37°C , and the extent of growth was used as a relative indicator of sensitivity . Since it is recommended that DTT-induced ER stress be performed in liquid rather than solid medium [26] , analysis of DTT sensitivity was performed by inoculating 10 , 000 conidia in liquid AMM containing the indicated concentrations of DTT and incubating at 37°C for 4 days . Utilization of skim milk was determined by inoculating 5 , 000 conidia onto skim milk agarose plates ( 0 . 5% skim milk , 0 . 8% agarose ) , or skim milk agarose supplemented with 1 . 2 M sorbitol . The plates were incubated at 37°C and radial growth was monitored daily for 3 days . For experiments involving S . cerevisiae , overnight cultures of the wt and the Δhac1 mutant ( Invitrogen ) were diluted to an OD600 of 0 . 1 and cultured at 30°C and 250 rpm until the OD600 reached 0 . 5 . Serial 5-fold dilutions were then spotted onto YPD ( 1% yeast extract , 2% peptone , 2% glucose ) plates containing TM ( 62 . 5 ng/ml ) or CFW ( 25–50 µg/ml ) , and the plates were incubated at 30°C or at 37°C for 2 days . Antifungal susceptibility of A . fumigatus strains was determined using the Etest diffusion assay ( AB BIODISK ) according to the manufacturer's instructions . Briefly , conidial suspensions were prepared in sterile distilled water and adjusted to 1×106 conidia/ml . One ml of the conidial suspension was then spread evenly onto the surface of a 150 mm plate of RPMI agar buffered with MOPS ( Remel , Lenexa , Kansas ) , using a glass spreader . The inoculated agar surface was allowed to dry for 20 min before Etest strips containing amphotericin B , caspofungin , itraconazole , fluconazole or ketoconazole were applied . The plates were incubated at 37°C for 48 h before being photographed . The MICs were read as the lowest drug concentrations at which the border of the elliptical inhibition zone intercepted the scale on the antifungal strip . For Etest experiments involving S . cerevisiae strains , overnight cultures in YPD were diluted to an OD600 of 0 . 1 and cultured at 30°C /250 rpm until the OD600 reached 0 . 5 . The cultures were diluted to an OD600 of 0 . 257 and the yeast were spread evenly onto the surface of duplicate YPD plates using a Q-tip . Etest strips were applied after allowing the plates to dry for 20 min , and the plates were incubated for 48 h at 30°C . All PCR primers used in the study are listed in Table 3 . The A . fumigatus hacA gene ( Genbank accession XM_743634 ) was disrupted using the split-marker approach [84] . The left arm of hacA was PCR amplified from genomic DNA using PFU turbo polymerase ( Stratagene ) with primers 522 and 523 creating PCR product #1 . The first two thirds of the hygromycin resistance cassette was amplified from plasmid pAN7-1 using primers 395 and 398 to create PCR product #3 . PCR products #1 and #3 were then combined in an overlap PCR reaction with primers 395 and 522 to generate PCR product #5 . PCR product #5 was then cloned into pCR-Blunt II- TOPO ( Invitrogen ) to create p527 . The right arm of the hacA gene was then PCR amplified from genomic DNA using primers 524 and 525 to generate PCR product #2 . The second two-thirds of the hygromycin resistance cassette was amplified from pAN7-1 with primers 396 and 399 to make PCR product #4 , and PCR products #2 and #4 were combined in an overlap PCR reaction with primers 396 and 525 to generate PCR product #6 . PCR product #6 was then cloned into pCR-Blunt II- TOPO to create p528 . The inserts from p527 and p528 were gel-purified following digestion with BstX1 and Smal , and 10 µg of each was used to transform wt-ΔakuA protoplasts as previously described [85] . Inositol was included into the selection plates since S . cerevisiae UPR mutants are inositol auxotrophs [34] . Hygromycin-resistant colonies were screened by PCR , and loss of the hacA gene was confirmed on monoconidial isolates by genomic Southern blot analysis as described in the results section . Probe A was PCR-amplified from wt genomic DNA using primers using primers 492 and 493 while probe B was amplified with primers 522 and 523 . Genomic Southern blot genotyping confirmed single-copy deletion of the hacA gene in 2 transformants , and these clones were used for phenotypic analysis . To construct the complementation plasmid , a phleomycin resistance cassette was excised from plasmid pBCphleo as a SalI/HindIII fragment and ligated into the XhoI/HindIII sites of plasmid pSL1180 . The trpC terminator was PCR-amplified from plasmid pAN7-1 and cloned into the vector as a SacII/SacI fragment to create plasmid pTPP . The hacA gene containing 1128 bp upstream of the predicted translational start site was PCR-amplified from wt genomic DNA using primers 522 and 527 and cloned into pCR-Blunt II-TOPO . The hacA gene fragment was excised from pCR-Blunt II- TOPO with a BamHI and NotI restriction digest and inserted into the BglII/NotI sites of pTPP . Ten µg of the plasmid was linearized with ApaI and transformed into ΔhacA mutant protoplasts as previously described [85] . Southern blot analysis of phleomycin-resistant colonies revealed that the complementation plasmid integrated homologously at the hacA locus , and at least one ectopic site ( data not shown ) . Total RNA was extracted from overnight cultures by crushing the mycelium in liquid nitrogen and resuspending in TRI reagent LS ( Molecular Research Center , Cincinnati , OH ) . The RNA was fractionated by formaldehyde gel electrophoresis and ribosomal RNA ( rRNA ) loading was visualized by SYBR-Green II staining and quantified using a STORM phosphorimager ( Molecular Dynamics ) . The RNA was transferred to BioBond nylon membranes ( Sigma ) , and hybridized to a 32P-labeled DNA probe for A . fumigatus bipA , pdiA , tigA or hacA . The bipA fragment was PCR-amplified from wt A . fumigatus genomic DNA using primers 494 and 495 , the pdiA fragment was amplified using primers 602 and 603 , the tigA fragment was amplified using primers 628 and 629 , and the hacA fragment was amplified with primers 492 and 493 . Hybridization intensities were quantified by Phosphorimager analysis and normalized against SYBR®-Green II-stained rRNA intensity . The uninduced form of the hacA mRNA ( hacAu ) was obtained by extracting RNA from overnight cultures of A . fumigatus , and the induced form of the hacA mRNA ( hacAi ) was obtained by extracting RNA from overnight cultures that were treated for 1 h with 1 mM dithiothreitol ( DTT ) . Confirmation that these conditions differentially modulated the conversion of hacAu to hacAi was obtained by Northern blot analysis prior to reverse transcription ( Figure 3A ) . The RNA was then reverse-transcribed using the Superscript III reverse transcriptase first-strand synthesis system ( Invitrogen ) and primers 493 and 572 . The resulting cDNAs were cloned into pCR-Blunt II-TOPO® and sequenced . Protease secretion in A . fumigatus was quantified using Azocoll ( Calbiochem ) hydrolysis as previously described [10] . Azocoll is an insoluble collagen linked to a red azo dye , and the release of the dye is indicative of collagen hydrolysis . Conidia were inoculated at a concentration of 1×105/ml in 50 ml of AMM-FBS ( Aspergillus minimal medium containing 10% heat-inactivated fetal bovine serum ( FBS ) as the nitrogen and carbon source ) [10] . The cultures were incubated at 37°C with gentle shaking at 150 rpm . After 72 h , a 1ml aliquot of the culture was microfuged at 15 , 000g for 5 min , and a 15 µl aliquot of the supernatant was added to 2 . 4 ml of a 5 mg/ml suspension of pre-washed Azocoll ( prepared by washing and resuspending the collagen particles in buffer containing 50 mM Tris ( pH 7 . 5 ) , 1mM CaCl2 , and 0 . 01% sodium azide ) . The collagen/supernatant mixture was incubated at 37°C for 3 h , with constant shaking at 350 rpm . The Azocoll/supernatant mixture was centrifuged at 13 , 000 g for 5 min and the release of the azo dye was determined by measuring the absorbance at 520 nm . Values were normalized to the lyophilized weight of the 72 h biomass . For analysis of total protein secretion , 2 . 5×107 conidia were inoculated into a 500 ml flask containing 100 ml of AMM and incubated for 3 d at 37°C without shaking . Under these conditions , the wt and ΔhacA strains generated a similar amount of dried biomass . The supernatant was removed from both cultures and concentrated to approximately 500 µl using the Amicon 8050 ultrafiltration system with a membrane cut-off of 10 kD . An equal volume of water was added to each sample and further concentrated to 100 µl using an Amicon Ultra Centrifugal Filter Device . Protein concentrations were determined using the Bradford assay and samples were stored at −80°C prior to analysis . For 1D gel analysis , each sample was mixed with sample buffer ( 200 mM Tris-HCl pH 6 . 8 , 50% glycerol , 5% sodium docecyl sulfate ( SDS ) , 0 . 5% bromophenol blue and 5% ( v/v ) β-mercaptoethanol ) , heated to 95°C and 5 µl of each sample was loaded onto a 12% SDS PAGE gel . Gels were run at 150 V for ∼2 h . After fixing for 30 min in 10% methanol and 7% acetic acid , gels were stained overnight with SYPRO Ruby fluorescent dye , destained for 30 min in fixing solution , then washed with water for 5–10 min prior to imaging with a GE Healcare Typhoon scanner . Mycelial cell wall fractionation was performed according to the method described by Fontaine et al . [86] with slight modification . Briefly , wt and Δhac1 strains were grown in a 1 . 2-liter fermenter in liquid Sabouraud medium . After 24 h of cultivation ( linear growth phase ) , the mycelia were collected by filtration , washed extensively with water and disrupted in a Dyno-mill ( W . A . Bachofen AG , Basel , Switzerland ) cell homogenizer using 0 . 5-mm glass beads at 4°C . The disrupted mycelial suspension was centrifuged ( 3 , 000×g for 10 min ) , and the cell wall fraction ( pellet ) obtained was washed three times with water , subsequently boiled in 50 mM Tris-HCl buffer ( pH 7 . 5 ) containing 50 mM EDTA , 2% SDS and 40 mM β-mercaptoethanol ( β-ME ) for 15 min , twice . The sediment obtained after centrifugation ( 3 , 000×g , 10 min ) was washed five times with water and then incubated in 1 M NaOH containing 0 . 5 M NaBH4 at 65°C for 1 h , twice . The insoluble pellet obtained upon centrifugation of this alkali treated sample ( 3 , 000×g , 10 min , AI-fraction ) was washed with water to neutrality , while the supernatant ( AS-fraction ) was neutralized and dialyzed against water . Both fractions were freeze-dried and stored at −20°C until further use . Hexose composition in the samples were estimated by gas-liquid chromatography using a Perichrom PR2100 Instrument ( Perichrom , Saulx-les-Chartreux , France ) equipped with flame ionization detector ( FID ) and fused silica capillary column ( 30 m×0 . 32 mm id ) filled with BP1 , using meso-inositol as the internal standard . Derivatized hexoses ( alditol acetates ) were obtained after hydrolysis ( 4N trifluoroacetic acid/8N hydrochloric acid , 100°C , 4 h ) , reduction and peracetylation . Monosaccharide composition ( percent ) was calculated from the peak areas with respect to that of the internal standard . For the cortisone acetate ( CA ) and cyclophosphamide ( CPS ) immunosuppression models , cultures were grown for 14 d on IMA agar ( Difco ) at 25°C . Conidia were collected by washing the agar surface with phosphate buffered saline ( PBS ) containing 0 . 05% Tween 20 . The conidial suspension was filtered first through sterile gauze and then through a 12 µm filter ( Millipore ) before washing twice with PBS/Tween . Female 4–6 week old C57BL/6 mice were used in all experiments , with the exception of the triamcinolone model . Inoculum sizes were selected on the basis of pilot experiments with the different immunosuppression methods to determine the minimum number of wt ( ΔakuA ) conidia that resulted in 100% mortality ( not shown ) . Mice were immunosuppressed with CA ( 2 mg subcutaneously ) administered on days −4 , −2 , 0 +2 and +4 in relation to infection , anesthetized with ketamine and xylazine and infected intratracheally with a target inoculum of 106 conidia for wt or ΔhacA ( n = 13 ) in PBS with 0 . 05% Tween 20 . Five sham-infected mice were immune suppressed and then inoculated intratracheally with PBS containing 0 . 05 % Tween 20 . Based upon plating efficiencies , mice received 1 . 1×106 of the wt or 1 . 2×106 conidia of the ΔhacA strains . Statistical significance was assessed by the log rank test using Sigma Stat 3 . 5 For the neutropenic model , mice were immunosuppressed with CPS ( 150 mg/kg intraperitoneally ) and monoclonal antibody RB6-8C5 ( 25 µg intraperitoneally ) one day before infection . CPS was then readministered three days after infection . Mice were infected intratracheally with a target inoculum of 5×105 conidia of wt or ΔhacA ( n = 13 or 12 per group , respectively ) in PBS with 0 . 05% Tween 20 . Five sham-infected mice were immunosuppressed and then inoculated intratracheally with PBS containing 0 . 05 % Tween 20 . Based upon plating efficiencies using the infecting inoculum , mice received 4 . 3×105 of the wt or 5 . 2×105 conidia of the ΔhacA strains . Statistical significance was assessed by the log rank test using Sigma Stat 3 . 5 . For the triamcinolone ( TA ) immunosuppression model , conidia were harvested from plates of AMM supplemented with 1 . 2 M sorbitol and resuspended in sterile saline . Groups of 10 CF-1 outbred female mice ( 20–28 g ) were immunosuppressed with a single dose of TA ( 40 mg kg−1 of body weight injected subcutaneously ) on day −1 . The mice were anaesthetized with 3 . 5% isofluorane and inoculated intranasally with 2×106 conidia from the wt , ΔhacA mutant , or the complemented strain on day 0 in a 20 µl suspension . Mortality was monitored for 15 days , and statistical significance was assessed by ANOVA using Sigma Stat 3 . 5 . | The pathogenic mold Aspergillus fumigatus is the leading cause of airborne fungal infections in immunocompromised patients . The fungus normally resides in compost , an environment that challenges the organism to obtain nutrients by degrading complex organic polymers . This is accomplished by secreted enzymes , some of which may also contribute to nutrient acquisition during infection . Extracellular enzymes are folded in the endoplasmic reticulum ( ER ) prior to secretion . If the folding capacity of the ER is overwhelmed by increased secretory demand , the resulting ER stress triggers an adaptive response termed the unfolded protein response ( UPR ) . In this study , we uncover a previously unknown function for the master transcriptional regulator of the UPR , HacA , in fungal virulence . In the absence of HacA , A . fumigatus was unable to secrete high levels of proteins and had reduced virulence in mice . In addition , loss of HacA caused a cell wall defect and increased susceptibility to two major classes of antifungal drugs used for the treatment of aspergillosis . These findings demonstrate that A . fumigatus relies on HacA for growth in the host environment and suggest that therapeutic targeting of the UPR could have merit against A . fumigatus , as well as other eukaryotic pathogens with highly developed secretory systems . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"cell",
"biology/microbial",
"physiology",
"and",
"metabolism",
"cell",
"biology/cellular",
"death",
"and",
"stress",
"responses",
"cell",
"biology/cell",
"signaling",
"infectious",
"diseases/fungal",
"infections",
"microbiology/microbial",
"growth",
"and",
"development",
"microbiology/medical",
"microbiology"
] | 2009 | A Role for the Unfolded Protein Response (UPR) in Virulence and Antifungal Susceptibility in Aspergillus fumigatus |
Reactome and WikiPathways are two of the most popular freely available databases for biological pathways . Reactome pathways are centrally curated with periodic input from selected domain experts . WikiPathways is a community-based platform where pathways are created and continually curated by any interested party . The nascent collaboration between WikiPathways and Reactome illustrates the mutual benefits of combining these two approaches . We created a format converter that converts Reactome pathways to the GPML format used in WikiPathways . In addition , we developed the ComplexViz plugin for PathVisio which simplifies looking up complex components . The plugin can also score the complexes on a pathway based on a user defined criterion . This score can then be visualized on the complex nodes using the visualization options provided by the plugin . Using the merged collection of curated and converted Reactome pathways , we demonstrate improved pathway coverage of relevant biological processes for the analysis of a previously described polycystic ovary syndrome gene expression dataset . Additionally , this conversion allows researchers to visualize their data on Reactome pathways using PathVisio’s advanced data visualization functionalities . WikiPathways benefits from the dedicated focus and attention provided to the content converted from Reactome and the wealth of semantic information about interactions . Reactome in turn benefits from the continuous community curation available on WikiPathways . The research community at large benefits from the availability of a larger set of pathways for analysis in PathVisio and Cytoscape . The pathway statistics results obtained from PathVisio are significantly better when using a larger set of candidate pathways for analysis . The conversion serves as a general model for integration of multiple pathway resources developed using different approaches .
Pathway diagrams are a common way to represent a wealth of information about biological molecules , interactions and processes . Currently , the Pathguide collection lists 45 freely available pathway databases with human data , out of which only 14 provide the data in a machine readable format [1 , 2] . Even fewer of these provide a pathway diagram that can be used for data visualization and downloaded for further analysis and conversion into other formats ( S1 Table ) . Notable among them are WikiPathways and Reactome , each with its unique user base , contributors , and curation cycle [3] . WikiPathways is an open , collaborative platform for drawing , curating , and sharing biological pathways , built using the same MediaWiki software underlying Wikipedia . WikiPathways leverages community curation to grow and maintain its pathway collection beyond the capabilities of an internal curation team . Anybody can register at WikiPathways to create new pathways and curate existing ones . WikiPathways provides a JavaScript-based viewer for interactively navigating and highlighting pathway elements and a Java based editor for creating and curating pathways . It makes use of BridgeDb web services [4] to provide identifier resolution and links to primary data sources . Pathways can be tagged for classification and quality control , e . g . pathways with the tag “curated” are regularly checked by a dedicated curation team and are deemed suitably annotated for analysis [5] . Pathways can also be tagged with various ontology tags from various pre-existing established ontologies , such as the Pathway ontology [6] and Disease Ontology [7] . Pathways from WikiPathways can be used to integrate , visualize , and analyze system-wide transcriptomics , proteomics , and metabolomics measurements using the open source pathway analysis tool PathVisio [8] . Pathways can also be analyzed as networks in Cytoscape [9] , using the WikiPathways app to convert the pathways into networks [10] . WikiPathways pathways are also used by several other tools , such as GO-Elite [11] and SNPLogic [12] . Domain experts often curate specific subsets of pathways in WikiPathways , which are made available in portals e . g . the plant portal [13 , 14] , CIRM portal [15] , exRNA portal [16] . In addition , WikiPathways data is available in RDF ( Resource Description Framework ) format , which is incorporated into the Open PHACTS Discovery platform , which integrates pharmacological data from a variety of information resources and provides tools and services to question this integrated data to support pharmacological research [17 , 18] . Like WikiPathways , Reactome is an open-source , open access pathway database with a substantial collection of diverse pathway models [19 , 20] . However , it differs from WikiPathways as the pathway annotations are annotations are curated by the Reactome editorial staff in collaboration with external experts in the research community . Reactome provides an intuitive website to navigate pathway knowledge and a suite of data analysis tools to support the pathway-based analysis of complex experimental and computational data sets . Similar to WikiPathways , visualization of Reactome pathways is facilitated by the Pathway Browser that supports zooming , scrolling , and highlighting , and can show detailed information about entities in the pathway . It makes use of PSICQUIC web services [21] to overlay molecular interaction data from the Reactome Functional Interaction Network [22] and external interaction databases , including IntAct [23] , and ChEBI [24] . Pathways in Reactome are explicitly constructed in terms of biochemical reactions and drawn in accordance with the community standard Systems Biology Graphical Notation ( SBGN ) [25] . Reactome also provides pathway analysis tools which can be used to perform ID mapping , pathway assignment , and over-representation or enrichment analysis with user-supplied datasets . The integration of Reactome content in WikiPathways provides Reactome with the power of community curation and broader format availability , including the semantic format using the WikiPathways RDF generator [26] . At the same time , WikiPathways benefits from the additional content and curation attention from the Reactome team . A connection between Reactome and WikiPathways was first proposed in 2008 , using either the EBI created CSV format and a novel converter , or the BioPAX format and Cytoscape [27] . However , neither of these routes was very successful in preventing loss of data . Therefore , these generic methods of conversion were abandoned for a more specific format conversion . Pathways in WikiPathways are stored using the Graphical Pathway Markup Language ( GPML ) format , while pathways in Reactome are stored in a relational database organized by the Reactome data model with their diagrams stored in the database as XML strings with other related information [28 , 29] . We created a converter to convert pathways directly from the relational database into the GPML format . In this manuscript , we describe the newly developed format converter to convert Reactome content for inclusion in WikiPathways . The addition of the Reactome pathways to the analysis collection of pathways available from WikiPathways improves the coverage of gene ontology biological process terms of the analysis collection to 90% . The converted Reactome pathways can be analyzed with several new analysis tools , such as the pathway analysis tool PathVisio and network analysis tool Cytoscape . As a pedagogic example , we perform pathway analysis using a publicly available transcriptomics dataset and the combined collection of pathways from WikiPathways and Reactome .
A pathway in WikiPathways consists of data nodes , interactions , and graphical elements , e . g . cellular compartments . Data nodes can be of the following types: gene product , protein , RNA , metabolite , pathway , complex , and unknown . Gene product is the default data node that can be used for all products of genes such as transcripts , proteins , RNAs , and genes . By default , these are represented as open rectangular boxes with black labels and borders . The more specific data node types such as protein and RNA can be used in the specific cases instead of a generic gene product node . The protein node is visually the same as the gene product node while RNAs are represented in purple . The metabolite node represents metabolites , drugs , or other small molecules; it is represented in blue . The pathway data node is used to denote a connection to another pathway , and represented in green without a border . The complex data node represents two types of complexes either a set of proteins represented as a brown rounded rectangle or a set of interacting proteins represented by a brown hexagon . Data nodes of type unknown are represented the same as the generic gene product node . Interactions describe the relationship between two data nodes . Currently , two collections of interactions are available in the drawing palette: basic interactions and Molecular Interaction Map ( MIM ) interactions [31] . Arrows can be used to describe basic interactions like conversion , translocation , activation , binding , and modification . T-bars denote inhibition . The MIM interaction palette can be used for more formal and easier machine-readable descriptions of Binding , Conversion , Catalysis , Stimulation and Necessary Stimulation , and Transcription/Translation . Graphical elements can be used to provide contextual meaning to the pathways . Graphical Shapes , lines , and labels can for instance be used to annotate a biological process and generally to make things visually clearer to biologists . Similarly , graphical cellular compartments such as mitochondria , endoplasmic reticulum , nucleus and cell walls can be also added to the pathway as predefined shapes for a richer diagram . Reactome uses a comparable but graphically slightly different method to describe pathway content . In Reactome , the core unit of the data model is the reaction . Entities ( nucleic acids , proteins , complexes , and small molecules ) participate in reactions . These reactions form a network of biological interactions and are grouped into pathways . Reactome uses the SBGN Process Description format [17] to draw pathway diagrams . Small Molecules are represented by a green oval , proteins by a green rounded rectangle , and complexes by blue hexagons . A group of entities playing the same roles in a reaction is annotated as EntitySet in Reactome , which is displayed as rounded rectangle with a double line border . Organelles are represented by orange rectangles with double line borders for membranes or single line borders for non-membrane organelles . The following reaction types can be represented: Transition/Process , Association/Binding , Catalysis , Inhibition , Dissociation , Omitted , and Uncertain . Stoichiometry , catalysis , positive and negative regulation , and other types of reaction attributes can also be represented in pathway diagrams based on SGBN . Fig 1 shows how the pathway elements from the Reactome pathways were represented in the converted pathway . Each element of the pathway can be annotated using database identifiers for data analysis and also annotated with literature references . As an example of the conversion , the Abacavir transport and metabolism pathway is shown here ( Fig 2 ) , example of a larger pathway is provided ( S1 Fig ) . In addition to converting the elements of the Reactome pathway diagram , the converter also draws the components of the complexes and entity sets at the bottom of the pathway . This helps with data visualization and gives better results for pathway analysis . Because all the complex and entity set members are also present in the same pathway diagram , they are also taken into consideration by the pathway statistics algorithm for determining the importance of the pathway for the given dataset in the given condition . The ComplexViz plugin enables the user to highlight the components on the bottom of the pathway belonging to the complex selected in the pathway diagram or vice versa . The Reactome converter was also used to convert plant pathways from the Plant Reactome database freely available at http://plantreactome . gramene . org/ . Pathways for the species Oryza sativa , Zea mays , and Arabidopsis thaliana were converted . The pathways for rice are manually curated , the pathways for the other species are computationally inferred from the rice pathways . These pathways have been made available in the plant portal at WikiPathways [14] . The biological entities in pathways and their relationships can be represented as nodes and edges in abstract biological networks . This opens up a large variety of network analysis methods to further extend , analyze and visualize biological pathways . The incorporation of the WikiPathways and ReactomeFIViz apps in the Cytoscape framework allows further investigation of biological pathways using a wide variety of Cytoscape apps for network analysis and visualization . The visualization of an example Reactome pathway with both apps is provided ( S3 Fig ) .
The WikiPathways project has developed a suite of pathway visualization and editing tools for users to view and edit pathways , and established a dynamic community to continuously crowd source updates and novel pathway content . The contents in Reactome are created by select domain experts in target fields of research with Reactome editorial staff . Including Reactome content has significantly expanded the coverage of pathway information at WikiPathways . Likewise , incorporating community edits from the WikiPathways versions of Reactome content significantly expands their pool of contributors , helping them produce more frequent updates and create links to outside databases . In the current implementation , we use a notification mechanism developed in WikiPathways to send edits from WikiPathways to Reactome . However , such an approach cannot be scaled up if many edits occur in the WikiPathways web site . We plan to develop a robust round-trip software approach in the Reactome curator tool so that edits in WikiPathways for Reactome pathways can be imported into the Reactome database easily . Such a tool will find new edits , and then present them to Reactome curators in graphical user interfaces so that curators can decide whether or not these edits should be committed into the Reactome database . We believe a true round-trip approach between Reactome and WikiPathways will benefit both projects , and set an example for other projects to collaborate with each other . The conversion of Reactome pathways to the GPML format enables the analysis and data visualization of Reactome pathways in PathVisio . PathVisio is a widely used pathway analysis software , preferred due to its excellent data visualization capabilities as demonstrated by its use in numerous academic publications [45–49] . PathVisio allows multiple data points to be visualized on one node using colors and color gradients permitting easy visualization of time series data . The data visualized images can then be exported as images for further publication or in html format as a mini-website to easily maneuver the uploaded data on the pathway image . The new ComplexViz plugin simplifies analysis of the converted Reactome pathways . As Reactome pathways typically contain numerous complexes , the plugin enables highlighting complex components on the bottom of the pathway diagram and the other way around . It also enables browsing complex components in a side panel and visualizing data uploaded for the complex components on the parent complex node . This highlights the complexes of interest , which can then be further studied . The complex component diagram on the side panel also displays the data uploaded thereby making it simpler to look at them without having to look for them on the bottom of the pathway . The pathway analysis case study presented here with a transcriptomics dataset comparing women with normal ovulatory physiology with those with PCOS shows that addition of the Reactome pathway set clearly improves pathway analysis results . The list of top ten most affected pathways feature pathways from both the curated and reactome_approved collection of pathways from WikiPathways . More pathways appear from the reactome_approved collection , which is expected since the collection is manually curated . The reactome_approved collection adds 4417 new gene products and 500 new metabolites . However , the curated collection still contains 1414 unique gene products and 2360 unique metabolites . There are 3438 gene products and 325 metabolites in common between the two collections . Therefore , the conversion adds content without much overlap . The toll like receptor ( TLR ) cascades pathway , which shows up as the most changed pathway in this condition is from the Reactome collection . TLRs are an important family of pattern recognition receptors ( PRR ) involved in innate immunity . The innate immune system initiates an inflammatory response after recognizing pathogens by PRRs [50] . Emerging evidence suggests that PCOS is associated with systemic inflammation [51 , 52] . Furthermore , various studies have reported that TLRs are expressed in the female reproductive tract[53] . Therefore , this pathway is clearly interesting for PCOS . In addition , the Cell surface interactions at the vascular wall pathway , which is the second most highly affected pathway is also from the Reactome collection . This pathway is annotated with the Gene Ontology biology process term , leukocyte migration . Since PCOS is associated with elevated levels of circulating leukocytes [54] , this pathway is clearly of interest . Both pathways , are from the newly converted collection of pathways from Reactome and clearly add biological knowledge , as illustrated with the case study for PCOS . The availability of Reactome pathways in WikiPathways allows the analysis of the pathways with several new analysis tools . Besides the analysis of pathways in PathVisio , users can also use the WikiPathways app for Cytoscape to analyze the pathways as biological networks . While the ReactomeFIViz app focuses on functional gene interaction networks , the WikiPathways app creates a representative network of the pathway including metabolites and other pathway elements . Consequently the created network provides a new analysis tool that allows the integrated analysis of different omics datasets .
A Java based format converter has been developed to convert pathways from the Reactome database to the WikiPathways format . Pathways in Reactome are stored in a relational database with their diagrams encoded in XML strings , while pathways in WikiPathways are stored as GPML , which is an XML based file format . In addition , both repositories have Java based APIs according to which the pathway files can be read and written . These internal data models of the two databases are used to read and write the pathways obtained from them . This allows the converter to remain flexible and backwards compatible as long as the data models themselves are . This also makes the converter stable through version updates of pathways as long the pathways are organized according to the same model . The conversion is done in the following steps: ( i ) Creating a GPML pathway and adding pathway attributes , ( ii ) Converting the pathway elements , and ( iii ) Annotating the pathway and pathway elements . These steps are described further below . The converter is open source and the code is available from the GitHub repository [55] . Human GO Terms were downloaded from UniProt-GOA [57] . Scripts in Java were written to parse the document to obtain the GO identifiers and identifiers of the terms for the three structured ontologies that describe gene products in terms of their associated biological processes , cellular components and molecular functions . The current release version 3 . 6 of HMDB was downloaded to obtain the superset of all human metabolites . Additional scripts in Java were written to map all gene products in the two pathway collections to Ensembl and all metabolites to HMDB . All the scripts used are available from GitHub [58] . The Ensembl gene identifiers were mapped to GO terms using Ensembl BioMart [59] , to obtain the total GO term coverage of the two pathway collections and also individual coverage of each GO category . The R package gplots was then used to create Venn diagrams showing GO and HMDB coverage [60] . The Venn diagrams were manually updated in PowerPoint . The newly developed plugin improves visualization of data on complexes and their components . The plugin can be installed in PathVisio using the plugin manager and adds a side panel “Components” . The top half of this panel displays the components of the complex that is clicked as a mini pathway diagram . Imported data is visualized both on the main pathway diagram and on the “Components” side panel containing the complex component diagram . Clicking on the buttons in the side panel next to the mini pathway diagram , displays the cross-references and expression data available for that data node on the bottom half of the panel . The plugin also adds the submenu item “Complex Visualization” to the Data menu . Clicking it opens a dialog box for setting visualization options for complexes . Three visualization options have been implemented . These methods allow changing the border color of complex and components , coloring complex nodes according to a calculated ratio , and drawing the complex label . Users can select a border color for complexes and their components to indicate which complex and components belong together . Complexes can be colored based on the percentage of complex components that qualify the user defined criterion . This percentage is calculated for all complexes on the pathway . Color gradients or rules can be used to visualize the score on the complexes . Text labels can be drawn on the Complexes after data has been visualized , the font and size of text of the label can be changed . The plugin is open source and the code is available from the GitHub repository [61] . A detailed user guide is provided ( S2 Text ) . An up-to-date copy will be maintained at the GitHub wiki [62] . | Biological pathways are descriptive diagrams that describe biological processes , i . e . interactions between genes , proteins , and metabolites . Pathways can therefore be used to integrate and visualize molecular measurements of genes , proteins , and metabolites in different biological conditions , e . g . healthy state vs . diseased state . This helps researchers investigate a disease . For instance , the low expression of a certain gene might in turn lead to the low abundance of a certain protein which might prevent the breakdown of a certain metabolite , the accumulation of which contributes to disease progression . High throughput “omics” technologies produce vast quantities of biological measurement data . Biological pathways provide an intuitive knowledge-based scaffold for integrating these data WikiPathways and Reactome are two commonly used pathway databases . Reactome pathways are centrally curated with periodic input by domain experts , while WikiPathways is a community-based platform where pathways are created and continually curated by any interested party . As part of an ongoing collaboration between Reactome and WikiPathways , we have added the Reactome pathways to WikiPathways and made them available from the Reactome portal on WikiPathways . Here , we demonstrate how such an integration is advantageous to both the Reactome and WikiPathways communities and to the general research community at large . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] | [
"medicine",
"and",
"health",
"sciences",
"protein",
"metabolism",
"engineering",
"and",
"technology",
"metabolic",
"processes",
"cancers",
"and",
"neoplasms",
"oncology",
"industrial",
"engineering",
"genome",
"analysis",
"quality",
"control",
"network",
"analysis",
"polycystic",
"ovary",
"syndrome",
"computer",
"and",
"information",
"sciences",
"metabolic",
"pathways",
"gynecological",
"tumors",
"gene",
"ontologies",
"biochemistry",
"data",
"visualization",
"genetics",
"biology",
"and",
"life",
"sciences",
"genomics",
"metabolism",
"computational",
"biology"
] | 2016 | Reactome from a WikiPathways Perspective |
Encapsidation is a strategy almost universally employed by viruses to protect their genomes from degradation and from innate immune sensors . We show that TRIM21 , which targets antibody-opsonized virions for proteasomal destruction , circumvents this protection , enabling the rapid detection and degradation of viral genomes before their replication . TRIM21 triggers an initial wave of cytokine transcription that is antibody , rather than pathogen , driven . This early response is augmented by a second transcriptional program , determined by the nature of the infecting virus . In this second response , TRIM21-induced exposure of the viral genome promotes sensing of DNA and RNA viruses by cGAS and RIG-I . This mechanism allows early detection of an infection event and drives an inflammatory response in mice within hours of viral challenge .
TRIM21 is a ubiquitously expressed high-affinity cytosolic antibody receptor and E3 ubiquitin ligase [1] . TRIM21 intercepts incoming antibody-opsonized virions during cellular infection , mediating efficient post-entry neutralization [2] and innate immune signaling [3 , 4] . Unlike Fc gamma receptors , which phagocytose immune complexes , TRIM21 detects antibody-bound virions that enter the cytosol after attachment of the virus to its specific cellular receptor , endocytosis , and endosomal escape . TRIM21 therefore detects viruses during what could otherwise be a productive infectious event and protects cells of diverse tissue types [3] . TRIM21 activation does not require any pathogen associated molecular pattern ( PAMPs ) or pattern recognition receptors ( PRRs ) but is based solely on sensing antibodies in the cytosol , an environment from which they are normally excluded . Consequently , TRIM21 is activated during infection by diverse pathogens including non-enveloped viruses and intracellular bacteria [3] . TRIM21 participates in both naïve infection ( through its ability to bind IgM ) and secondary infection ( by binding IgG ) . Upon in vivo challenge with mouse adenovirus 1 ( MAV-1 ) , mice lacking TRIM21 succumb to fatal viral infection within 7 days [5] . Upon detection of an antibody-coated virus in the cytosol , TRIM21 synthesizes K63 ubiquitin chains and activates the innate immune pathways NFκB , AP-1 and IRF3/5/7[3] . This leads to a broad program of antiviral transcription and induction of an anti-viral state . Concurrent with stimulating signaling , TRIM21 recruits p97/VCP , a AAA+ ATPase with segregase and unfoldase activity , and the proteasome , resulting in premature disassembly and degradation of viral capsids . This rapid degradation of incoming viral particles provides a potent block to infection [1 , 6] . We hypothesized that this catastrophic uncoating might also expose viral genomes for sensing by nucleic acid pattern recognition receptors ( PRRs ) . Despite the presence of numerous cytosolic PRRs that recognize pathogen nucleic acids , it is often viral transcripts or progeny genomes , as opposed to incoming genomes , that are detected [7 , 8 , 9] . Here we show that TRIM21 potentiates the sensing of antibody-bound DNA or RNA viruses by cytosolic nucleic acid sensors cGAS or RIG-I . In addition , we identify that TRIM21-dependent innate immune signaling contributes a substantial component of the antibody block to rhinovirus infection in cultured cells , and demonstrate that TRIM21 and neutralizing antibody together drive a rapid pro-inflammatory transcriptional response upon adenovirus infection in mice .
Recombinant human adenovirus type 5 ( AdV ) dose-dependently activates immune transduction pathways and this can be sufficient to upregulate transcription of cytokines and chemokines [10] . We tested whether immune activation is potentiated in the presence of antibody ( IgG ) . Activation of an NFκB reporter was observed during infection of AdV+IgG at a multiplicity of infection ( MOI ) > 2 , which increased until an MOI of 100 ( Fig 1A ) . In contrast , NFκB activation during infection of AdV alone was only observed at an MOI >100 , although induced TNFα transcription could be detected at slighter lower viral doses ( Fig 1B ) . Similar antibody potentiation of NFκB activation was observed during infection by the picornavirus human rhinovirus type 14 ( HRV ) , although both NFκB and TNFα could be observed at a high viral dose ( Fig 1C and 1D ) . Picornaviruses antagonize immune sensing through expression of 3C protease , which cleaves immune signalling components [11 , 12] . Previously , we have shown that complement C3 promotes RhV sensing but is a substrate for 3C protease[13] . Treatment with rupintrivir , which inhibits cleavage , increases complement C3-dependent immune activation . In contrast , we found that rupintrivir did not alter IgG-dependent sensing ( Fig 1E ) . To confirm that 3C protease is capable of suppressing immune activation in our experiments , we over-expressed the protease and found that IgG-mediated upregulation of TNFα was reduced ( Fig 1F ) . Taken together , this suggests that IgG potentiates sensing of rhinovirus infection in a manner that is not antagonized by 3C protease . This is consistent with rapid TRIM21-dependent degradation of incoming antibody-coated virions prior to their replication . Picornaviruses use diverse mechanisms to infect cells[14] . While the major group virus HRV14 is believed to disrupt endosomal membranes similar to AdV , leading to lysis and release of the virion into the cytosol , others like the minor group virus HRV2 are thought to form a pore through which only the genome is passed[14] . To test whether the observed antibody-dependent enhancement of immune activation requires delivery of antibody-opsonized virions to the cytosol we compared NFκB activation during HRV2 and HRV14 infection . While HRV14 induced a robust NFκB response in the presence of antibody , HRV2 did not ( Fig 1G ) . Importantly , immune activation was observed when HRV2+IgG was transfected into cells using PEI ( polyethylenimine ) , an endosomal disruptor[15] , supporting the hypothesis that exposure to the cytosol is a key determinant of sensing in these experiments . To investigate the kinetics of antibody and TRIM21 dependent immune transcription , mouse embryonic fibroblasts ( MEF ) cells were challenged with AdV or HRV at a viral dose where no immune activation is observed in the absence of antibody ( an MOI of 40 or TCID50 of 250 , respectively ) . In the presence of antibody , infection provoked two sequential waves of TNFα transcription ( Fig 2A and 2D ) . Comparison with TRIM21-/- MEF cells showed that TRIM21 is required for both the earlier ( 4 hours , Fig 2B and 2E ) and later ( 8 hours , Fig 2C and 2E ) responses , but not for TNFα transcription induced by transfected poly ( I:C ) . We confirmed that , like for adenovirus , genetic deletion of TRIM21 had no impact on HRV infection or replication per se , but that TRIM21 was required for efficient neutralization of HRV ( S1A–S1C Fig ) . Continuous media exchange to remove secreted cytokines from the cell supernatant had relatively little impact on TNFα induced by AdV+IgG compared to transfected DNA ( Fig 2F ) , suggesting TRIM21-dependent signaling at 8 hours is the result of a second , independent , sensing event . Comparing a panel of immune transcripts induced by AdV+IgG or HRV+IgG supports the interpretation that the second wave of signalling is not just an amplification of the first wave . At 4 hours post infection , AdV+IgG stimulated transcription more potently than HRV+IgG whereas at 8 hours this pattern was not maintained ( Fig 3 ) . In fact , there was substantially greater transcription of ISGs 8 hours after infection with HRV+IgG compared to AdV+IgG , suggesting that additional sensing events have occurred that differ between the two viruses . There was little change in any transcript upon infection in the absence of antibody , consistent with the NFκB and TNFα induction data ( Fig 1 and S2 Fig ) . We hypothesized that while the first wave of signaling may correspond to TRIM21 synthesis of K63-chain ubiquitin upon antibody binding , as previously described [3] , the second wave may be the result of TRIM21-dependent exposure of viral PAMPs . To test this , we measured immune activation after transfection of cells with antibody-coated latex beads . Antibody-coated beads but not beads alone induced NFκB in a TRIM21-dependent manner ( Fig 4A ) , consistent with previously published data[3] . Importantly , while antibody-coated beads activated TNFα transcription 4 hours post-transfection , there was no measurable response at 8 hours ( Fig 4B ) . This is consistent with the second wave of sensing requiring a viral PAMP . TRIM21 mediates rapid capsid degradation of incoming viruses , suggesting that the viral PAMP stimulating the second wave of transcription could be exposed viral genomes . Whilst we did not observe sensing of incoming AdV alone , equivalent concentrations of purified AdV DNA induced a TNFα response in transfected cells , suggesting that AdV DNA is stimulatory if exposed ( Fig 4C ) . To provide evidence that the viral genome is the PAMP revealed by TRIM21 in the second sensing wave , we treated virus with ultraviolet ( UV ) irradiation to crosslink the nucleic acid . UV-crosslinking had no impact on antibody-dependent signaling at 4 hours post infection but TNFα transcription at 8 hours was substantially reduced ( Fig 4D and 4E ) . Adenovirus is one of the most resistant pathogens to UV crosslinking[16] , but comparison to untreated virus confirmed that irradiation had sufficiently modified the DNA to prevent it from undergoing replication ( Fig 4F ) . Taken together , this suggests that unmodified genomes are required for the second TRIM21-dependent sensing wave . Paraformaldehyde ( PFA ) is a chemical crosslinker that covalently couples protein-protein complexes together and has been used to stabilise virions and prevent their disassembly [17 , 18] . We hypothesised that if TRIM21-mediated virion degradation is required to expose the genome and provoke the second sensing wave then this should be inhibited by PFA-treatment . Consistent with this , PFA-crosslinking of adenovirus before incubation with antibody prevented TNFα transcription at 8 hours but not 4 hours post-infection ( Fig 4G ) . To confirm that PFA crosslinking stabilises adenovirus virions we measured the impact of treatment on the uncoating of EdU-labelled virus . There was a significant decrease in the proportion of uncoated genomes ( defined as dissociated from hexon ) after PFA treatment ( Fig 4H and 4I ) . Genome uncoating is a pre-requisite for nuclear entry and we also observed that PFA-crosslinking prevented import of vDNA into the nucleus ( Fig 4J ) . Kutluay et al . have previously shown that TRIM5α-mediated proteasomal degradation of incoming retroviral particles results in exposure of viral nucleic acid [19] . We used a similar approach to support our hypothesis that TRIM21-mediated proteasomal degradation also exposes viral genomes . Using qPCR to track incoming viral genomes , we noted that incubation with a non-entry blocking but TRIM21-dependent anti-adenovirus antibody ( wt9C12 ) [2] did not substantially reduce adenovirus uptake ( similar levels were observed at 1hr ) but led to a marked reduction in AdV DNA after 2hrs ( Fig 4K ) . Furthermore , 9C12 increased the rate of genome loss between 2–6 hours compared to infection by virus alone , a measure that is independent of how many genomes have entered the cell . The presence of nucleases in the cytosol capable of degrading exposed DNA was confirmed by incubating lysate with a FRET-labeled DNA probe ( S3 Fig ) . Mutant antibody with reduced TRIM21 binding ( mut9C12 ) or proteasome inhibition ( epoxomicin ) decreased genome loss but not to levels observed with virus alone ( Fig 4L and 4M ) , consistent with previously published data showing that these perturbations reduce but do not abolish TRIM21 activity[1 , 2] . Importantly , while mut9C12 reduces TRIM21 binding it does not prevent binding to hexon ( S4 Fig ) . To correlate the inhibitory effect of PFA-crosslinking on second wave sensing with prevention of genome exposure , we measured the impact of 9C12 on genome loss after infection with treated virus . Unlike with untreated virus , we observed no reduction in genome copies in the presence of wt9C12 ( Fig 4N ) , suggesting that TRIM21 cannot mediate the efficient degradation of a PFA-crosslinked virion . 5-ethylyne-2’-deoxyuridin ( EdU ) in-situ labelling of the viral genome uses ‘click’ chemistry to install a small fluorescent probe on modified uracil nucleotides . It therefore may not distinguish uncoated nucleoprotein cores from naked vDNA , whose accessibility to DNA sensors may differ[20] . In contrast , bromodeoxyuridine ( BrdU ) labelling relies on the access of antibody to modified uracil bases and is therefore thought to be inhibited by proteins bound to viral DNA [21] . We used AdV with bromodeoxyuridine ( BrdU ) -labeled DNA ( BrdU-AdV ) and an anti-BrdU antibody to determine whether antibody-induced uncoating increases accessibility of the viral genome . While accessible AdV DNA was rarely detected following infection with BrdU-AdV alone ( Fig 5A and 5D ) , pre-incubation of BrdU-AdV with wt9C12 antibody ( Fig 4B and 4D ) , but not mut9C12 ( Fig 5C and 5D ) , increased the appearance of BrdU-positive puncta . Furthermore , overexpressed FLAG-cGAS , a cytosolic DNA sensor , localized to these antibody-dependent BrdU-positive puncta ( Fig 5E ) . The fact that far fewer virions were detected using BrdU labelling than in our EdU labelling experiments ( Fig 4H ) is consistent with the former requiring more accessible vDNA and may explain why adenovirus is poorly sensed in the absence of antibody even though uncoated genomes are present in the cytosol . Taken together , these data suggest that TRIM21-mediated capsid uncoating exposes the viral genome , allowing detection by cytosolic nucleic acid PRRs . To test whether the genomes of incoming virions are sensed by nucleic acid PRRs in a TRIM21-dependent manner , we used siRNA to deplete MEF cells of STING or MAVS , the convergent adaptor proteins required for sensing cytosolic DNA and RNA , respectively[22] ( S5A Fig ) . Depletion was confirmed by challenging cells with transfected DNA or poly ( I:C ) ( S5B Fig ) . Neither STING nor MAVS depletion reduced antibody-dependent signaling 4 hours post infection with either AdV+IgG or HRV+IgG ( Fig 6A ) , in agreement with previous results[3] and confirming that the first wave of sensing is primarily dependent on TRIM21 detection of antibody delivery to the cytosol . However , at 8 hours post infection , depletion of STING partially reduced TNFα transcription induced by AdV+IgG , while depletion of MAVS reduced transcription upon infection with either AdV+IgG or HRV+IgG ( Fig 6B ) . These results are consistent with the known MAVS-dependent sensing of HRV RNA[23 , 24 , 25]; and detection of AdV by STING-dependent DNA sensors[10 , 26] and RNA polymerase III-dependent MAVS activation[7] . Although the later pathway involves host transcription of adenoviral DNA , it can occur prior to nuclear entry and productive infection , and in response to transfected poly ( dA-dT ) DNA , thus can still be considered a mechanism of detection of incoming nucleic acids[7] . MAVS and STING have been shown to transduce signals down the TBK1/IRF3 and IKK/NFkB [27 , 28] pathways . To investigate the involvement of these pathways in detection of AdV+Ab we used 5Z-7-Oxozeaenol , which inhibits TAK1 , panepoxydone , which prevents IκB degradation , and the TBK1 inhibitor BX795 , which blocks the phosphorylation and nuclear translocation of IRF3[29] . 5Z-7-Oxozeaenol and panepoxydone both inhibited NFκB induction 7 hours post-infection with AdV+Ab ( Fig 6C ) . Moreover , NFκB inhibition had a direct effect on AdV+Ab-induced cytokine transcription . Transcripts of CXCL10 , TNFα , CCL2 , CCL4 , IFIT1 & IL6 during the second wave were all reduced by either NFκB inhibitor ( Fig 6D ) . Addition of BX795 inhibited both the first and second waves of TNFα transcription comparably with panepoxydone ( Fig 6E and 6F ) . Together , this data suggests that immune activation induced by infection in the presence of antibody utilizes canonical transduction pathways known to be downstream of MAVS and STING . Upstream receptors that activate MAVS and STING include the RNA sensor RIG-I and the DNA sensor cGAS . Consistent with the MAVS knockdown , depletion of RIG-I ( S5C and S5D Fig ) reduced TNFα activation by AdV+IgG and HRV+IgG at 8 hours ( Fig 6H ) , but not 4 hours ( Fig 6G ) . Meanwhile , siRNA depletion of cGAS ( S5E and S5F Fig ) inhibited detection of AdV+IgG 8 hours ( Fig 6J ) , but not 4 hours , post infection ( Fig 6I ) . Repeating our experiments using interferon priming to increase cGAS levels modestly beyond those under conditions of AdV+IgG or HRV+IgG infection at 4 hours post infection ( Fig 6K and 6L ) had no impact on sensing at either 4 hours ( Fig 6M ) or 8 hours ( Fig 6N ) . Thus second wave sensing is not due simply to upregulated sensors but is consistent with antibody-enhanced detection of viral genomes . To determine the physiological importance of TRIM21-mediated immune activation , we assessed its contribution to controlling HRV infection . In cells stably depleted of TRIM21 , neutralization of HRV infection was substantially reduced ( Fig 7A and 7B ) . Treatment of TRIM21-depleted cells with NFκB signaling pathway inhibitors , panepoxydone ( PPD ) and 5Z-7-oxozeaenol ( 5Z7O ) , gave no additional reduction in antibody protection . NFκB inhibition in control cells resulted in antibody protection intermediate between untreated control cells and TRIM21-depleted cells , suggesting that signaling forms part of the antiviral activity of TRIM21 , with the remaining component due to TRIM21-dependent neutralization . Finally , we tested whether the ability of TRIM21 to promote immune activation prior to viral replication facilitates a rapid inflammatory response during infection in vivo . We infected TRIM21+/+ or TRIM21-/- mice with mouse adenovirus type 1 ( MAV-1 ) by intraperitoneal injection in the presence or absence of anti-adenovirus immune serum ( serum , S ) . A rapid inflammatory response was observed 6 hours post infection in the brain tissue of TRIM21+/+ animals that also received immune serum . In contrast , animals deficient in TRIM21 or lacking immune serum had significantly reduced or absent induction of IFNα ( Fig 7C ) , IFNβ1 ( Fig 7D ) , TNFα ( Fig 7E ) and the interferon stimulated gene IRF7 ( Fig 7F ) .
The early detection of infection is advantageous to the host . However , most viruses are shielded by a protein capsid that protects their genomes and associated PAMPs from immune sensing . As a consequence , incoming viral genomes are not always detected by pattern recognition receptors and detection often requires progeny genomes produced at later stages of the infectious cycle [30] . For instance , the capsid of HIV-1 mediates a series of post-entry events that allows the virus to evade detection in macrophages by the cytosolic PRR cGAS[31 , 32] . Similarly , adenovirus infection proceeds via an elegant and highly coordinated stepwise process of disassembly[33] , that may function in part to minimize exposure of the genome to DNA sensors in the cytoplasm . Here we show that TRIM21 , which mediates the rapid degradation of incoming viral capsids in the cytosol , promotes the ability of sensors cGAS and RIG-I to detect the genomes of infectious viruses . Upon infection with either the DNA virus adenovirus or the RNA virus rhinovirus we observed two early waves of innate immune signaling , both of which were dependent on TRIM21 . The first wave led to a relatively similar transcription pattern of immune-responsive genes for both viruses and did not require the adaptors MAVS or STING or the sensors RIG-I or cGAS . In contrast , the second wave of transcription was variously dependent upon these adaptors and sensors and led to a divergent enhancement of the response between viruses . In the absence of TRIM21 , there was no significant immune induction within the first 8 hours of infection by either AdV or HRV . This was despite the fact that transfection of AdV vDNA provoked robust TNFα induction when assayed at 8 hours . Thus TRIM21 not only promotes the early detection of diverse viruses but also potentiates the activity of other PRRs to detect the highly immunostimulatory ligands that are shielded in an infectious viral particle ( Fig 8 ) . Adenovirus virions can be sensed in the absence of antibody but at higher MOIs . Why adenovirus is normally poorly sensed is an important question , particularly given recent data showing that most uncoated viral genomes are not imported into the nucleus[21] . Our BrdU-labelling experiments suggest that although these genomes are dissociated from capsid , the vDNA is poorly accessible . This is consistent with the densely packed structure of the protein VII-containing adenovirus nucleoprotein core and its known insensitivity to nucleases and DNA repair enzymes[20 , 34] . Viral nucleoprotein complexes may be as important as the capsid shell in protecting incoming viral genomes from sensors . The ability to intercept incoming virions allows TRIM21 to trigger an immune response during the early stages of cellular infection . We detected robust transcription of type I interferons , cytokines and ISGs 6 hours post infection in the brain tissue of wild-type but not TRIM21 mice challenged with MAV-1 . An important aspect of TRIM21 function is that it depends on antibodies to detect virions in the cytosol . Consistent with this , we did not observe significant early immune activation in mice in the absence of immune serum . Importantly , these results identify a previously unrecognized component of antibody-mediated protective immunity in vivo . Furthermore , they provide an unusual example of adaptive immunity activating innate immunity , contrary to the classic progression from innate to adaptive responses via pro-inflammatory mediators . IgM antibodies of the primary repertoire , which TRIM21 can utilize , may be capable of inducing an early response during naïve infection , however we did not observe this during our experiments with MAV-1 . While TRIM21-mediated disassembly of infecting virions promotes both viral restriction and exposure of PAMPs , there may be competition between sensing and nuclease-mediated degradation of the genome . Indeed , this may contribute to the duration of TRIM21-dependent signaling waves . Whether genome degradation is a necessary step in viral restriction has not been fully investigated , but the premature uncoating of adenovirus may be sufficient to reduce infection by preventing microtubule engagement and trafficking of virions to the nucleus . This is supported by data showing that incubation of AdV with 9C12 inhibits localization of virions to the nuclear envelope [35] . TRIM21 degradation of incoming virus may be advantageous to sensing not only in revealing PAMPs but also in preventing the expression of viral antagonists of immune signaling . The sensing of progeny genomes is known to be highly susceptible to viral antagonism . For instance , influenza nonstructural protein 1 is a very effective RIG-I antagonist but is only produced upon replication [36] . The sensing of incoming virions and genomes by TRIM21 , together with cGAS and RIG-I , may be harder for the virus to inhibit . In support of this , we found that TRIM21-dependent signaling is not antagonized by virus-produced HRV protease 3C but is inhibited if the protease is ectopically overexpressed prior to infection . Furthermore , TRIM21 potentiates the sensing of HRV by RIG-I , which does not normally contribute to picornavirus sensing because it is cleaved and rendered inactive by 3C protease [11] . Antagonism by 3C may be one explanation for the lack of immune detection of HRV in our experiments in the absence of TRIM21 . TRIM21 may be facilitating sensing of HRV both because it exposes the incoming genome to RIG-I and because it degrades virions before 3C can be expressed . Given the similarity between the dual functions of TRIM21 and those of TRIM5 as a retroviral capsid sensor and restriction factor[37 , 38 , 39] , our findings suggest that TRIM5 may also promote detection of retroviral infection by revealing the incoming genome or early reverse transcription products from within the viral capsid . More generally , this study highlights the ability of PRRs to act synergistically . This may be advantageous to the host , both in amplifying immunity , and in guiding the resulting response to best counter the invading pathogen .
TRIM21+/+ and TRIM21-/- MEF cell lines were generated as described previously1 . HeLa cells are a human epithelial cervix carcinoma cell line ( ATCC , Manassas , VA ) . Cells were maintained in Dulbecco’s modified eagle medium ( DMEM ) supplemented with 10% fetal bovine serum ( FBS ) and 100 U/ml penicillin and 100 μg/ml streptomycin . All cells were incubated at 37°C in the presence of 5% CO2 , except where stated . Knockdown by siRNA: MEF cells were transfected with siRNA using RNAiMAX ( Life Technologies , Carlsbad , CA ) , and were incubated for 2 days before use in assays . siRNA as follows: negative control ( Ambion , Carlsbad , CA ) , STING ( Sc-154411 , Santa Cruz , Dallas , TX ) , MAVS ( s105943 , Ambion ) , cGAS ( 1:1 mixture of duplexes with the following sequence: GAUUGAGCUACAAGAAUAU , GAGGAAAUCCGCUGAGUCA , Sigma ) , RIG-I ( L-065328 , Dharmacon , Lafayette , CO ) . Knockdown by shRNA: cells were transduced with retroviral particles encoding small hairpin RNA ( shRNA ) against human TRIM21 ( GCAGCACGCTTGACAATGA ) as described previously [2] . Human adenovirus type 5 vector ( ΔE1 , ΔE3 ) expressing GFP ( AdV ) was purchased from ViraQuest ( North Liberty , IA ) . EdU-AdV and BrdU-AdV was produced by infection of the transcomplementation cell line HEK293T in the presence of 7 . 5 μM 5’-ethynyl-2’-deoxyuridine ( EdU ) or 5 μg/ml BrdU respectively ( Sigma , St . Louis , MO ) . PFA-AdV was produced by incubation of AdV on ice for 3 hours in 4% PFA , followed by quenching with 0 . 1 M glycine , then dialysis to AdV storage buffer ( 25 mM Tris pH 8 , 0 . 5 M NaCl ) . AdV and HRV were UV-crosslinked by exposure to 500 mJ/cm2 UV irradiation using a UV Stratalinker 2400 ( Stratagene , La Jolla , CA ) . Human rhinovirus type 2 & 14 ( HRV ) were produced by infection of HeLa cells , and purified by 2 rounds of CsCl centrifugation as described previously [1] . Mouse adenovirus type 1 ( MAV-1 ) ( ATCC ) was produced by infection of a 3T3 MEF cell line ( ATCC ) , and virus was purified by 2 rounds of CsCl centrifugation as above . HRV replication: 5 x 103 cells were plated in 96 well plates overnight in DMEM supplemented with 2% FCS ( low-serum media ) . Where stated , recombinant human IFNα ( Sigma ) was added to cells at the time of plating . 20 TCID50 units HRV was mixed 1:1 with human serum IgG ( Sigma ) ( or PBS where no IgG was added ) for 1 hour then added to cells . Cells were incubated for 6 hours , then washed and incubated in fresh low-serum media at 35°C for 7 days . Viability was then measured using 500 μg/ml MTT ( 3- ( 4 , 5- Dimethylthiazol-2-yl ) -2 , 5-Diphenyltetrazolium Bromide ) reagent ( Sigma ) , with absorbance measurement at 540/650 nm using a SpectraMAX 340PC ( Molecular Devices , Sunnyvale , CA ) . AdV , UV-crosslinked AdV and PFA-crosslinked AdV were titrated onto HeLa cells ( 5 x 104 cells per well plated overnight ) then incubated for 24 hours . The percentage of cells positive for GFP expression was determined by fluorescence activated cell sorting ( FACS ) analysis using a BD LSRII machine . AdV neutralization: 3 . 75 x 104 IU AdV per well was mixed 1:1 with human serum IgG ( IgG ) at the stated concentration then added to HeLa cells ( 5 x 104 cells per well plated overnight , MOI = 0 . 3 ) and incubated for 24 hours , then the percentage of cells positive for GFP expression was determined as above . C57BL6 wild type ( TRIM21+/+ ) and TRIM21-/- mice used in all in vivo experiments were obtained from Jackson Laboratories ( Bar Harbor , ME ) . All mice used in experiments were between 7 to 10 weeks old at the start of infection protocols . All experiments were conducted in accordance with the 19 . b . 7 moderate severity limit protocol and the Home Office Animals ( Scientific Procedures ) Act ( 1986 ) . TRIM21+/+ and TRIM21-/- mice were infected by IP injection with 3 . 12 x 106 TCID50 units MAV-1 in 100 μl PBS . MAV-1 antiserum was prepared from blood collected from C57BL/6 mice 72 days post infection following 3 rounds of immunization with a sub-lethal dose of MAV-1 as described previously [5] . Mice were injected by IP with 200 μl serum diluted 1:1000 . In all cases , control mice received injection of the same volume of PBS buffer . At the end point of the experiment , brains were extracted and snap-frozen in liquid nitrogen . TRIM21+/+ and TRIM21-/- MEF cells were plated at a density of 1–2 x 104 per well in 96-well plates overnight prior to infection . Viruses were mixed 1:1 with antibody and incubated for 30 mins at room temperature prior to infection . AdV was used at 7 . 5 x 105 IU per well , HRV was used at 2 . 5 x 102 TCID50 units per well unless otherwise stated . Human serum IgG ( IgG ) was used at 2 . 5 mg/ml . Where stated , human IFNα ( 0 . 1 U per well , Sigma ) was added at the time of infection . p ( I:C ) ( 1 μg/ml , Sigma ) , herring sperm DNA ( 10 μg/ml , Life Technologies ) and p ( dA-dT ) ( 10 μg/ml Sigma ) , were all transfected using lipofectamine 2000 ( Life Technologies ) . AdV DNA was purified from virions using a DNeasy mini kit ( Qiagen , Manchester , UK ) , and also transfected using lipofectamine 2000 . Purified AdV DNA concentration and DNA concentration in AdV virions ( following 10 mins incubation in 0 . 1% SDS detergent to disassemble virions ) were determined using a NanoDrop 2000c ( Thermo Scientific , Loughborough , UK ) , so that the equivalent concentration of total AdV DNA could be applied to cells by either infection or transfection . In all experiments using transfection of nucleic acids , results are normalized to mock transfection containing lipofectamine 2000 but no nucleic acids ( however , this gave no significant signaling above baseline ) . Where indicated , cells were washed with PBS 1 hour post stimulation ( infection or transfection of nucleic acid ) , then cells were either incubated in fresh media for the remaining time or media was exchanged every 15 mins from this point until the end of the experiment . At the stated end timepoints , cells were washed with PBS , then cDNA was prepared using Taqman Cells-to-CT ( Fast ) kit ( Life Technologies ) . Relative gene expression compared to TBP reference gene was determined using the change-in-threshold ( 2−ΔΔCT ) method , using Taqman gene expression assays ( Life Technologies ) on a StepOnePlus Real Time PCR machine ( Applied Biosystems , Waltham , MA ) . Taqman gene expression assay mixes as follows: TATA-Box binding protein ( TBP ) Mm00446971_m1 , Tumor necrosis factor α ( TNFα ) Mm00443260_g1 , Interferon α4 ( IFNα4 ) Mm00833969_s1 , Interferon β1 ( IFNβ1 ) Mm00439552_s1 , CXCL10 Mm00445235_m1 , Interferon regulatory factor 7 ( IRF7 ) Mm00516791_g1 , CCL5 Mm01302427_m1 , Myxovirus resistance 1 ( Mx1 ) Mm00487796_m1 , MAVS Mm00523170_m1 , STING Mm01158117_m1 , RIG-I Mm01216853_m1 , cGAS Mm00557693_m1 . MEF cells were plated at a density of 1 x 104 cells per well in 96-well plates overnight before infection with AdV . AdV was used at 7 . 5 x 105 infectious units ( IU ) per well mixed 1:1 with wt9C12 and mut9C12 ( both 25 μg/ml , produced as described previously ( McEwan et al . 2013 ) ) in a total volume of 5 μl for 30 mins before addition to cells . 1 hour post infection , cells were washed with PBS x 3 then incubated in fresh media until the endpoint of the experiment . Where PFA-AdV was used , the equivalent of 1 . 5 x 106 infectious units ( IU ) per well mixed 1:1 with wt9C12 and mut9C12 at 50 μg/ml . AdV DNA was measured from cell lysates produced using lysis buffer ( without DNase ) from the Taqman Cells-to-CT ( Fast ) kit ( Life Technologies ) , followed by qPCR for GFP ( primers and probe as previously described [2] ) , using a GFP plasmid standard curve for copy number determination . To compare GFP genome copy number between untreated and PFA-crosslinked virus we used standard curves created from serial dilution of equivalent volumes of AdV and PFA-AdV ( S6 Fig ) . Where stated , 5Z-7-oxozeaenol ( 0 . 5 μM , Sigma ) , panepoxydone ( 2 μg/ml , Enzo Life Sciences , Farmingdale , NY ) , epoxomicin ( 2 μM , Sigma ) , or DMSO ( Sigma ) solvent control were added to cells 1 hour prior to infection . MEF cells were infected with BrdU-AdV or BrdU-AdV pre-incubated with antibody ( as above ) for 2 hours . HeLa cells were transfected with FLAG-cGAS ( human , OriGene , Rockville , MD ) using FuGENE 6 ( Promega , Brentwood , UK ) 24 hours , before infection as above . Cells were fixed in 4% paraformaldehyde and permeabilized with 0 . 01% digitonin ( Sigma ) , and non-specific binding was blocked using 5% BSA . BrdU was visualized using BU1/75 ( Life Technologies ) and AlexaFluor 488-anti-rat ( Life Technologies ) . EdU genome labelling was performed using Click-iT EdU AlexaFluor 488 imaging kit ( ThermoFisher Scientific ) according to manufacturer’s instructions before staining with AlexaFluor 568-conjugated 9C12 . FLAG-cGAS was visualized using M2 anti-FLAG ( Sigma ) and AlexaFluor 568-anti-mouse . Samples were mounted in Vectashield with DAPI ( Vector Laboratories , Peterborough , UK ) , and imaged using an LSM 780 microscope ( Carl Zeiss MicroImaging , Oberkochen , Germany ) . Cell lysates were electrophoresed on Nu-PAGE gels ( Life Technologies ) and transferred to nitrocellulose membranes using the iBlot rapid transfer system ( Life Technologies ) . Antibodies to human TRIM21 ( Sc-25351 , Santa Cruz ) with HRP-conjugated anti-mouse ( A0168 , Sigma ) , and HRP-conjugated beta-Actin ( Sc-47778 , Santa Cruz ) were used to detect protein levels . The DNase activity assay was adapted from the DNase Alert quality control ( QC ) checking kit ( Life Technologies ) . Cell lysate was prepared from 1 x 105 cells resuspended in hypotonic buffer ( 20 mM HEPES pH7 . 5 , 5 mM NaF , 10 μM Na2MoO4 and 100 μM EDTA ) and incubated on ice for 15 minutes before addition of 50 μl 10% Nonidet P-40 ( New England Biolabs ( UK ) , Ipswich , UK ) and vortexing for 10 seconds . Lysates were spun at 13000 xg for 30 seconds at 4°C and the supernatant was collected as the cytoplasmic fraction . Cytosolic extract was mixed with NucleaseAlert buffer and DNA oligonucleotide probe , and cleavage kinetics were monitored using a BMG Pherastar FS platereader at 530/580 nm . Relative fluorescence was calculated by subtracting fluorescence from that achieved by hypotonic buffer and Nonidet P-40 treated controls . 80 mg sections of forebrain were cut and homogenised in 1 ml Qiazol ( Qiagen ) using a TissueRuptor ( Qiagen ) . Total RNA was purified using the RNeasy lipid tissue kit ( Qiagen ) with on-column DNase I digestion kit ( Qiagen ) according to the manufacturer’s protocol . cDNA was generated from 1 μg purified RNA per reaction , using SuperScript III ( 200 U per reaction , Life Technologies ) and oligo ( dT ) 23 primers ( 7 μM final concentration , Sigma ) . Relative gene expression was determined as for cDNA from cultured cells ( see above ) . Data was graphed using Graphpad Prism 6 software ( Graphpad Software , La Jolla , CA ) . Error bars show standard errors of the mean ( SEM ) . Statistical significance was determined using unpaired two-tailed student’s T-tests , also using Graphpad Prism 6 software . Mouse infection was carried out under project license 80/2534 , which was approved by the Medical Research Council Animal Welfare and Ethical Review Body . All experiments adhered to the 19 . b . 7 moderate severity limit protocol as per the UK Home Office Animals ( Scientific Procedures ) Act ( 1986 ) . | Our cells have potent immune sensors that can detect the presence of viral nucleic acid in the cytosol . Unfortunately , almost all viruses utilize a strategy of encapsidation , comprising a protein shell that protects their genomes and impedes them from being sensed or degraded . In our study , we describe how components of innate and adaptive immunity combine to allow the rapid sensing of genomes from incoming viruses . We show that a ubiquitous immune protein called TRIM21 intercepts virions immediately after they enter the cytosol and exposes their genomes to nucleic acid sensors , thereby activating immune transcription pathways before genome replication commences . We demonstrate that TRIM21 enables the RNA sensor RIG-I to detect infection by an incoming RNA virus and the DNA sensor cGAS to detect infection by a DNA virus . By facilitating the sensing of incoming rather than progeny genomes , TRIM21 facilitates a rapid immune response upon infection . In the final part of our manuscript , we illustrate that this system confers an advantage to the host in vivo by demonstrating that there is a rapid TRIM21-dependent inflammatory response in mice upon viral infection , whereas in the absence of TRIM21 production of crucial cytokines like interferon is delayed . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Material",
"and",
"Methods"
] | [] | 2015 | TRIM21 Promotes cGAS and RIG-I Sensing of Viral Genomes during Infection by Antibody-Opsonized Virus |
Cytopathic effects ( CPEs ) in mosquito cells are generally trivial compared to those that occur in mammalian cells , which usually end up undergoing apoptosis during dengue virus ( DENV ) infection . However , oxidative stress was detected in both types of infected cells . Despite this , the survival of mosquito cells benefits from the upregulation of genes related to antioxidant defense , such as glutathione S transferase ( GST ) . A second defense system , i . e . , consisting of antiapoptotic effects , was also shown to play a role in protecting mosquito cells against DENV infection . This system is regulated by an inhibitor of apoptosis ( IAP ) that is an upstream regulator of caspases-9 and -3 . DENV-infected C6/36 cells with double knockdown of GST and the IAP showed a synergistic effect on activation of these two caspases , causing a higher rate of apoptosis ( >20% ) than those with knockdown of each single gene ( ∼10% ) . It seems that the IAP acts as a second line of defense with an additional effect on the survival of mosquito cells with DENV infection . Compared to mammalian cells , residual hydrogen peroxide in DENV-infected C6/36 cells may signal for upregulation of the IAP . This novel finding sheds light on virus/cell interactions and their coevolution that may elucidate how mosquitoes can be a vector of DENV and probably most other arboviruses in nature .
The dengue virus ( DENV ) , a flavivirus belonging to the family Flaviviridae , is the etiological agent of dengue fever ( DF ) , dengue hemorrhagic fever ( DHF ) , and dengue shock syndrome ( DSS ) [1] . The genome of the DENV consists of a single-stranded positive-sense RNA of ∼11 kilobases ( kb ) long; which possesses an m7G cap at the 5′-end and is non-polyadenylated at its 3′-end [2] . During infection within a host cell , viral RNA directly translates into a single polyprotein that is subsequently cleaved into three structural proteins and seven nonstructural proteins by the combined action of host proteases and the trypsin-like viral NS2B/NS3 serine protease [3] . The DENV and other viruses generally invade a host cell by redirecting cellular processes in order to meet the needs of viral propagation . This , in turn , leads to novel changes in the expressions of various genes [4]–[6] . During infection by the DENV , an unfolded protein response ( UPR ) , or so-called UPR signal cascade , is usually induced in a time-dependent manner [7]; it may be able to cope with endoplasmic reticular ( ER ) stress in host cells [8] . Moreover , specific serotypes of DENV may modulate the UPR with different selectivities [9] . Generally , the UPR can provide an early defensive mechanism for infected cells to survive ER stress due to viral infection by eliminating misfolded proteins and allowing cells to recover . [10] . In contrast , it might also modify the outcome to benefit viral replication [11] . Despite this , the cell type may also play a role in determining responses to viral infections , leading to differential fates of infected cells [12] . Arboviruses , e . g . , DENV , require arthropods as vectors for effective transmission in nature [13] , indicating that the viruses can efficiently replicate in both invertebrate and vertebrate cells . It seems that mosquitoes are the mixing tanks of arboviruses . Thus , co-evolution between mosquito cells and viruses is required . However , such viruses apparently cause different outcomes in different cell types [14] . It was reported that apoptosis ultimately occurs in most flavivirus-infected mammalian cells , even though the UPR is activated [15] . Nevertheless , minor cytopathic effects ( CPEs ) are commonly seen in DENV-infected mosquito cells [16] . In fact , mosquito cells after DENV infection generally overexpress antioxidant genes such as glutathione S transferase ( GST ) which copes with virus-induced oxidative stress , leading a high survival rate of infected cells [14] . Although this significant reduction in cell death may be mediated by disruption of redox signaling , it is , however , apparently not effective enough to save infected cells according to observations of changes in caspases and apoptosis rates between cells with and without DENV infection [14] . It is conceivable that trivial damage to infected cells is a prerequisite for a mosquito or arthropod to be capable of efficiently transmitting arboviruses in nature . As a result , it is worthwhile determining the occurrence of an antiapoptotic mechanism in mosquito cells in response to DENV infection . It is particularly important to identify what factors may be involved in regulating the occurrence of apoptosis and to investigate how it is induced during infection . We preliminarily identified an inhibitor of apoptosis ( IAP ) gene that is upregulated in mosquito cells with DENV infection . In this study , we tried to further demonstrate the role of the IAP in DENV-infected mosquito cells and annotate the link between antiapoptotic effects and antioxidant defense which was described previously [14] .
C6/36 cells derived from the mosquito Aedes albopictus were used for DENV-2 ( New Guinea C strain ) propagation; cells were cultured in minimal essential medium ( MEM; GIBCO™ , Invitrogen , Carlsbad , CA , USA ) supplemented with 10% fetal bovine serum ( FBS ) , 2% non-essential amino acids , 2 g/ml Hepes ( Sigma , St . Louis , MO , USA ) , 2 . 2 g/ml sodium bicarbonate ( NaHCO3 ) , and 0 . 4% of an antibiotic-antimycotic ( GIBCO™ , Invitrogen ) in a closed system at 28°C . The propagated virus was titrated as previously described in BHK-21 cells derived from baby hamster kidney [5] , which were maintained in MEM containing 10% FBS , 2% non-essential amino acids , 2 . 2 g/ml NaHCO3 , and 0 . 4% of an antibiotic-antimycotic in an atmosphere containing 5% CO2 at 37°C . C6/36 cells ( ∼107 cells/tube ) harvested in tubes were centrifuged at 3000 rpm and 4°C for 10 min . The DENV-2 suspension or fresh medium ( with mock infection used as the control ) was added to the tubes , after removing the medium , at a multiplicity of infection ( MOI ) of 1 for incubation at 28°C for 1 h with gentle agitation every 15 min . The viral suspension was then removed by centrifugation , and pelleted cells were seeded and incubated at 28°C for 24 h . RNA extraction and reverse-transcription polymerase chain reaction ( RT-PCR ) procedures were performed following a previous description [4] . Briefly , total RNA was isolated from both mock- and DENV-2-infected C6/36 cells using the Trizol reagent ( Invitrogen ) following the protocol in the manual provided by the manufacturer , for further experiments . The approach followed a protocol to generate a double-stranded ( ds ) -oligo as a tool of miR RNAi ( Invitrogen ) . The miR RNAi sequence to knock down the GST gene ( miR-GST ) was generated by annealing the top- and bottom-strand oligos containing the linkers ( top: 5′TGCTG… . 3′; bottom: 5′CCTG… . C3′ ) , the mature miR-GST antisense target sequence , the loop sequence , and the sense target sequence ( 5′cgtgatgtgcctggagggaat3′ ) , to form ds-oligos ( top strand: tgctgattccctccaggcacatcacggttttggccactgactgaccgtgatgtctggagggaat; bottom strand: cctgattccctccagacatcacggtcagtcagtggccaaaaccgtgatgtgcctggagggaatc ) . Construction of this system was described in our previous report [14] . Transcript changes in the GST gene in cells , either transfected or not , were validated by a real-time RT-PCR using forward ( GST-RTF: ACCGAGGATTATGCCAAGATG ) and reverse primers ( GST-RTR: TCGCACAAATACTGGAGGATG ) . The approach followed a protocol to generate a ds-oligo as a tool of miR RNAi ( Invitrogen ) . The miR RNAi sequence ( including 2 fragments of miIAP-1 and miIAP-2 ) to knock down the IAP gene was generated by annealing the top- and bottom-strand oligos containing the linkers ( top: 5′TGCTG…3′; bottom: 5′CCTG…C3′ ) , the mature miR-IAP , the antisense target sequence , the loop sequence , and the sense target sequence ( 5′TGACGTGTGAGGAAATGATCA3′ ) , to form ds-oligos ( top strand: TGCTGTGATCATTTCCTCACACGTCAGTTTTGGCCACTGACTGACTGACGTGTGGAAATGATCA; bottom strand: CCTGTGATCATTTCCACACGTCAGTCAGTCAGTGGCCAAAACTGACGTGTGAGGAAATGATCAC ) . Construction of the vector was performed as described previously [14] . Transcript changes in the IAP gene in cells , either transfected or not , were validated by a real-time RT-PCR using the forward ( IAP-RTF: ATCGCGAGAAGAGGAGCATC ) and reverse primers ( IAP-RTR: TGCCATCATCATTTGAGCCA ) . The primers used for IAP detection were derived from a sequence from the EST dataset previously established by our laboratory . The primer pair used for detection of IAP from BHK-21 cells by real time RT-PCR included the forward ( BHIAPRTF: GGAGGCGGTTAGACAGAAG ) and reverse primers ( BHIAPRTR: GTATAGCACAGGGTCACTCTC ) . After constructing the mi-IAP-silenced construct , we transiently transfected the miIAP construct into miGST-C6/36 cells . miGST-C6/36 cells were seeded into 6-well plates 12 h prior to transfection . We prepared the transfection mix reagent containing 100 µl of FBS and antibiotics-free medium , 2 µg of the miIAP construct , and 8 µl FuGENE® HD ( Roche , Berlin , Germany ) , which was then incubated for 15 min at room temperature . After 15 min , miGST-C6/36 cells were washed with antibiotics-free medium without FBS , and such medium was added to the transfection mix reagent to a final volume of 1 ml . We removed the transfection mix reagent from miGST-C6/36 cells , incubated the cells for 5 h , and then replaced the medium with serum-containing medium 24 h after transfection . Assays for caspases-9 and 3 activated in apoptosis cell followed the protocol of kits provided by the manufacturer ( BioVision , Mountain View , CA , USA ) , which were measured by flow cytometry using caspase inhibitors conjugated to sulfo-rhodamine . Both C6/36 and BHK-21 cells were used for the previously described assay [14] . The method for measuring cell death followed a previously reported description [14] . In brief , C6/36 cells ( ∼2×106 cells/tube ) with or without transfection of the miR-IAP were collected and infected with DENV-2 at an MOI of 1 . At 12 , 24 , 36 , and 48 hpi , cells ( both infected and uninfected ) were harvested for measurement by flow cytometry . The monolayer of C6/36 cells inoculated with the viral suspension ( at an MOI of 1 ) was absorbed for 1 h . After the fluid was removed , the antioxidant reagent , N-acetyl-L-cysteine ( L-NAC ) ( Sigma-Aldrich , St . Louis , MO , USA ) , at a concentration of 10 mM ( pH 7 . 4 ) adjusted with fresh medium , was added for further culture . Cells were harvested to measure H2O2 , caspases , and apoptosis , as described previously [14] . The monolayer of C6/36 cells ( with or without transfection of miR-GST ) infected with DENV-2 at an MOI of 1 in a Petri dish ( 10 cm in diameter ) was washed with phosphate-buffered saline ( PBS; pH 7 . 3 ) , and then treated with trypsin-EDTA for 5 min . One milliliter of PBS containing 10% FBS was added to the dish , and cells were incubated with 10 µM 2′ , 7′-dichlorofluorescein ( CM-H2DCFDA ) ( Sigma-Aldrich ) at 28°C in the dark for 30 min . Cells were then harvested and subjected to analysis by a fluorescence-activated cell sorter ( FACScalibur , Becton Dickinson , Immunofluorometry Systems , Mountain View , CA , USA ) with excitation at 535 nm and emission at 610 nm . An immunofluorescent assay was used to detect localization of double-stranded ( ds ) RNA or the envelope ( E ) protein of the DENV-2 . Infected and uninfected C6/36 cells were smeared on a cover glass which was then washed with PBS ( pH 7 . 4 ) three times . Subsequently , cells were fixed in 4% paraformaldehyde for 10 min , washed again with PBS , and then blocked with 1% bovine serum albumin ( BSA ) in PBS for 1 h at 37°C . Primary polyclonal antibodies ( 1∶100 in dilution ) of anti-dsRNA ( J2 ) ( English & Scientific Consulting , Szirak , Hungary ) were added to the cover glass and incubated at 37°C for 1 h . The cover glass was subsequently incubated with secondary antibodies ( 1∶100 ) conjugated with FITC at 37°C for 1 h after they had been washed with PBS . The cover glass was finally mounted with a mixture of glycerol and PBS ( 3∶7 ) and observed under a laser scanning confocal microscope ( Zeiss LSM 510 , Vertrieb , Germany ) . Negative controls were incubated with diluents without primary antibodies; otherwise samples were subjected to the same procedures described above . The antibody against protein E of the DENV-2 was kindly provided by Prof . C . L . Kao , Department of Clinical Laboratory Sciences and Medical Biotechnology , National Taiwan University ( Taipei , Taiwan ) . For electron microscopy , cells seeded on the dish or those scraped off the culture dish ( and centrifuged at 4°C and 3000 rpm for 10 min ) were immediately fixed with a mixture of 2% ( v/v ) glutaraldehyde and 2% paraformaldehyde in 0 . 1 M cacodylate buffer overnight at 4°C . After cells were post-fixed in 1% ( w/v ) osmium tetroxide in 0 . 1 M cacodylate buffer for 2 h at room temperature , they were washed with 0 . 2 M cacodylate buffer three times . Again , cells were washed with 0 . 2 M cacodylate buffer three times and then dehydrated through an ascending graded series of ethanol . Cells were finally embedded in Spurr's resin ( Electron Microscopy Science , Hatfield , PA , USA ) and polymerized at 70°C for 72 h . The trimmed blocks were sectioned with an ultramicrotome ( Reichert Ultracut R , Leica , Vienna , Austria ) . Ultrathin sections were sequentially stained with saturated uranyl acetate in 50% ethanol and 0 . 08% lead citrate . Selected images were observed and photographed under an electron microscope ( JEOL JEM-1230 , Tokyo , Japan ) at 100 kV . Comparisons between two means were analyzed by Student's t-test at a significance level of 5% .
In this study , we assessed changes in caspases and their effects on apoptosis occurring among DENV-infected mosquito cells . Our results revealed that no evident change in activated caspase-9 occurred in C6/36 cells even when they had been infected by the virus for 48 h ( Fig . 1A ) , while it had increased 3 . 3-fold at 48 h hpi in infected BHK-21 cells which usually ended up dead ( Fig . 1B ) . Similarly , activated caspase-3 remained at a low level ( <10 mean fluorescent intensity; mfi ) in both C6/36 and BHK-21 cells even after they had been infected with DENV for 24 h ( Fig . 1C ) . However , it had significantly increased to 3 . 4-fold in BHK-21 cells , but not C6/36 cells , at 48 hpi ( Fig . 1D ) . In DENV-infected C6/36 cells , the IAP gene was significantly 2 . 4-fold upregulated at 36 hpi , and had reached 3 . 8-fold by 48 hpi; however , it had decreased to 50% and 60% by 36 and 48 hpi , respectively , in infected BHK-21 cells ( Fig . 2 ) . This suggests that the IAP plays a role in this process . It implies that an antiapoptotic effect through the IAP serves as a second defense system , protecting cells from virus-induced apoptosis in C6/36 cells . In this study , we also applied an miRNA-based silencing technique to knock down the IAP in C6/36 cells , and obtained a 50% knockdown efficiency ( Fig . 3A ) . Through this knockdown system , the apoptosis rate was evaluated , and showed an increase of 8 . 3% in those cells at 48 hpi ( Fig . 3B ) . In the meantime , elevated ( 1 . 46-fold ) caspase-9 activity appeared in IAP-knockdown and DENV-infected C6/36 cells at 48 hpi ( Fig . 3C ) , while caspase-3 showed an increase of 1 . 54-fold at the same time after infection ( Fig . 3D ) . It was observed that caspases are significantly activated in BHK-21 cells infected with DENV , producing a high proportion of cell death through apoptosis . In contrast , C6/36 cells with DENV infection upregulated IAP expression which suppressed the activities of caspases-9 and -3 , leading to an antiapoptotic effect in C6/36 cells as shown above . We thus further doubly knocked down C6/36 cells to reduce expressions of the IAP and GST . Cells with DENV infection revealed that caspase-9 had increased 2 . 6-fold at 48 hpi , although it was at a low level at 24 hpi ( Fig . 4A ) . The pattern of caspase-3 activation was similar to that of caspase-9; its activity had increased 2 . 83-fold by 48 hpi ( Fig . 4B ) . These results showed a significant difference in caspase activities between singly and doubly knocked-down cells . The apoptosis rate had increased to 23 . 52% for doubly knocked-down C6/36 cells with DENV infection by 48 h , which significantly differed from that of cells with single knockdown of the GST gene ( Fig . 4C ) . These results revealed there are two separate lines of defense involved in reducing apoptosis in mosquito cells with DEVN infection: one operated by GST and the other by the IAP . In C6/36 cells infected with DENV at an MOI of 1 , dsRNA , representing active viral RNA replication , and E proteins , representing the synthesis of viral proteins , were frequently observed at 12 hpi ( Fig . 5A ) . In the meantime , a great number of progeny virions that formed as a crystalline array within membrane-bound vacuoles widely appeared , mostly at 24 hpi , in the cytoplasm ( Fig . 5B ) . It was noted that C6/36 cells were morphologically intact even though thousands of virions have formed in those cells . The present study measured H2O2 , by CM-H2DCFDA staining , produced in DENV-infected BHK-21 and C6/36 cells , which showed a significant difference between the two cell types . The former usually presented vigorous H2O2 elevation , basically beginning at 36 hpi and remaining through 48 hpi , while the latter revealed a slight increase at 36 hpi which had further increased by 48 hpi . However , C6/36 cells with GST knockdown presented a pattern of H2O2 production similar to that of BHK-21 cells ( Fig . 6 ) . This indicates that H2O2 increased much later and more slightly in mosquito cells in response to DENV infection , suggesting that a stable environment remained in C6/36 cells particularly in the early stage of infection . The sulfhydryl antioxidant , L-NAC ( 10 mM ) , was used to treat C6/36 cells with and those without GST-knockdown ( miGST ) in order to compensate for the antioxidant ability within cells . In GST-knockdown C6/36 cells , both caspases-9 and -3 were activated at 36∼48 hpi , while only minimal activation was apparent at 48 hpi in cells treated with L-NAC ( Fig . 7A ) . Quantitative results showed that caspase-9 had significantly increased by 36 hpi in GST-knockdown C6/36 cells after infection; however , it decreased after treatment with L-NAC ( Fig . 7B ) . Production of caspase-3 also showed a similar trend to that of caspase-9 ( Fig . 7C ) . Looking at the apoptosis rate in GST-knockdown C6/36 cells , L-NAC treatment obviously alleviated the apoptosis rate to a relatively low level ( 5 . 69% at 48 hpi ) ( Fig . 8A ) . In contrast , untreated cells had a higher apoptosis rate at 36 hpi ( 10 . 45% ) which had further increased to 11 . 08% by 48 hpi ( Fig . 8B ) . The apoptotic effect of L-NAC treatment on uninfected cells has been demonstrated to be extremely low ( Fig . 8C ) . Our observations showed that the antioxidant status is undoubtedly associated with elevated expression of the IAP which ultimately protects DENV-infected C6/36 cells . Treatment with L-NAC to strengthen the antioxidant ability , presumably by altering the antioxidant status , actually alleviated IAP expression in C6/36 cells ( Fig . 9 ) . In C6/36 cells without L-NAC treatment , induction of the IAP usually increased in response to DENV infection , particularly at 36 and 48 hpi ( Fig . 9 ) .
Many flaviviruses have been known to evolve specific tactics , resulting in evasion of the innate and adaptive immune responses [17]; which usually occur among mammalian hosts in response to infection of the viruses [18] , [19] . However , it seems like mosquitoes preferentially select innate responses due to lack of an adaptive immunity [20] . For instance , Toll , Imd ( immune deficiency ) , and JAK-STAT pathways are frequently implemented by insects including mosquitoes , those pathways usually play a role in limiting viral replication within host cells [21] . According to reported , flaviviriral infections of mosquito cells can also be modulated by the mosquito's RNA interference pathway [22] . In general , mosquito cells are unable to use a potent cellular response to eliminate viral infected cells as mammalian cells . It suggests that a compromising between host cells and viral replication may be the best choice for mosquito survival . In turn , it has recently been reported that mosquito cells may utilize antioxidant defense as one mechanism to protect themselves [14] . In the meantime , inducing or inhibiting apoptosis may also be involved in affecting arbovirus replication in mosquito cells [23] . We previously demonstrated that mitochondria are involved in the caspase cascade and resulting apoptosis in host cells with DENV infection [14] , likely via the so-called mitochondrial or intrinsic pathway . Since apoptosis is inevitable in DENV-infected BHK-21 cells , activation of both caspase-9 and -3 is apparently essential in those cells . Theoretically , activation of caspase-9 disrupts mitochondrial diffusion limits and leads to the further release of cytochrome ( Cyt ) c [24] . In turn , Cyt c-dependent formation of an Apaf-1/caspase-9 complex activates caspase-3 and eventually initiates an apoptotic protease cascade [25] . Since C6/36 cells usually survive DENV infection , it is thought that there must be a negative regulator or suppressor of apoptosis that protects C6/36 cells infected by the DENV . Apoptosis is important for a variety of life phenomena [26]; it can also be a cellular response to viral infections [27] . Caspases are central components of the machinery responsible for apoptosis and/or programmed death of cells [28]; they belong to a group of enzymes known as cysteine proteases [29] . It is now known that apoptosis is induced via either activation of death receptors ( the extrinsic pathway ) or mitochondria ( the intrinsic pathway ) . In most cases , both pathways converge to induce activation of caspases which are cleaved to form the active forms before apoptosis occurs [30] . Induction of apoptosis via death receptors typically results in activation of an initiator caspase such as caspase-8 or -10 , which then activate other caspases in a cascade , eventually leading to activation of effector caspases , such as caspases-3 and -6 [31] . On the other hand , the mitochondrial pathway of apoptosis begins with a change in the permeability of the mitochondrial outer membrane [32] . Generally , activation of caspase-9 ( the initial caspase ) followed by caspase-3 ( the effector caspase ) is mediated by the formation of an apoptosome , leading to the occurrence of apoptosis [33] . Caspases are normally suppressed by IAPs that bind to and inhibit active caspases [34] . IAPs are a family of proteins originally identified in baculoviruses [35] , which play roles in protecting infected cells from death [36] . In turn , IAPs act as regulators to inhibit activation of caspases and thus dysregulate cell death pathways [37] . The IAP identified from the mosquito Ae . albopictus is composed of 402 amino acids and contains two baculoviral IAP repeat ( BIR ) domains and a RING-finger domain at its carboxyl terminus [38] . Significant upregulation of the IAP gene in DENV-infected C6/36 cells suggests that IAPs play roles in protecting cells from virus-induced apoptosis [38] , [39] . It further implies that an antiapoptotic effect through IAPs serves as a second defense system , leading to the absence of apoptosis in C6/36 cells during DENV infection [36] . We applied an miRNA-based silencing technique to knock down the IAP in C6/36 cells in this study , and showing increased apoptosis and elevated activities of caspases-9 and -3 . These results completely coincided with the pathway of intrinsic apoptosis , and suggested that loss of the IAP makes mosquito cells highly sensitive to virus-induced oxidative stress , representing an important regulator of inhibition in nature [40] . We further doubly knocked down C6/36 cells to reduce the expressions of the IAP and GST and show significant differences in caspase activities between singly and doubly knocked-down cells . This suggests that mosquito cells have two strategies , one operated by GST and the other by the IAP , to cope with stresses by inducing genes associated with defense against stress . However , this greatly differs from antiviral strategies in mammalian cells [41] , a high proportion of which end up undergoing apoptosis . Multiple defense responses through activation of various defense-related genes were observed in Cucumber mosaic virus ( CMV ) -infected tomato plants [42] . In this study , we further confirmed an additional effect implemented by the IAP and GST , leading to enhanced protection of C6/36 cells against DENV infection . We have repeatedly observed that a high survival rate of mosquito cells with DENV infection . Although the antioxidant defense by GST partially contributes to the ability to protect cells , there is supposedly a second defense system to enhance or synergize protection of mosquito cells from DENV infection . It is believed that the antiapoptotic pathway of the IAP plays such a role in elevating protection of infected mosquito cells . Since H2O2 can act as a redox signal at a lower level within cells [43] , it was demonstrated to modulate downstream signaling events such as calcium mobilization , protein phosphorylation , and gene expression [44] . As the result shown above , only a slight amount of H2O2 was produced in mosquito cells with DENV infection . In the present study , the antioxidant , L-NAC , was used to treat C6/36 cells with GST-knockdown ( miGST ) in order to neutralize free radicals and consequently protect cells from oxidative stress [45] . Results revealed that antioxidant-deficient C6/36 cells could effectively stimulate IAP expression through recovery by L-NAC treatment . It seems that the antioxidant status can regulate expression of the IAP and the resultant cell fate . In other word , H2O2 may adjust itself to act as a key player to overcome stress-mediated apoptosis [46] , inducing the IAP which serves as a second defense pathway to enhance protection of C6/36 cells from DENV infection . Although arbovirus infection of mosquito cells may trigger cell death , it occurs only when virus replication exceeds a threshold level [21] . This reflects that conventional innate defense of mosquitoes may delay virus replication , leading to a prosperous status of virus growth at the later stage within infected cells . Overall , this study provided important evidences that mosquito cells can survive DENV infection via an antioxidant defense which is followed by an antiapoptotic effect . These two defense systems are linked by an appropriate dose of residual H2O2 which was reported to be critical for apoptosis inhibition induced by UV irradiation [47] . This exquisite defense network promotes cell survival of infected mosquitoes; trivial damage to infected cells may be a prerequisite for mosquitoes and arthropods serving as efficient transmitters of arboviruses in nature . | This study demonstrated an idea that mosquito cells can survive dengue virus ( or other arboviruses ) infection through antioxidant defense and an additional effect by induction of IAP expression for protection of infection . It makes mosquito eligible to support virus replication efficiently , leading to a goal which is important to explain how mosquitoes can be a vector even when they have been seriously infected by the virus . Our findings opened an avenue for studies on virus/vector co-evolution that benefits for both virus replication and its transmission to humans or susceptible hosts . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] | [
"medicine",
"infectious",
"diseases",
"biology",
"evolutionary",
"biology",
"viral",
"diseases",
"vectors",
"and",
"hosts"
] | 2012 | Additive Protection by Antioxidant and Apoptosis-Inhibiting Effects on Mosquito Cells with Dengue 2 Virus Infection |
Staphylococcus aureus RNAIII is the intracellular effector of the quorum sensing system that temporally controls a large number of virulence factors including exoproteins and cell-wall-associated proteins . Staphylocoagulase is one major virulence factor , which promotes clotting of human plasma . Like the major cell surface protein A , the expression of staphylocoagulase is strongly repressed by the quorum sensing system at the post-exponential growth phase . Here we used a combination of approaches in vivo and in vitro to analyze the mechanism used by RNAIII to regulate the expression of staphylocoagulase . Our data show that RNAIII represses the synthesis of the protein through a direct binding with the mRNA . Structure mapping shows that two distant regions of RNAIII interact with coa mRNA and that the mRNA harbors a conserved signature as found in other RNAIII-target mRNAs . The resulting complex is composed of an imperfect duplex masking the Shine-Dalgarno sequence of coa mRNA and of a loop-loop interaction occurring downstream in the coding region . The imperfect duplex is sufficient to prevent the formation of the ribosomal initiation complex and to repress the expression of a reporter gene in vivo . In addition , the double-strand-specific endoribonuclease III cleaves the two regions of the mRNA bound to RNAIII that may contribute to the degradation of the repressed mRNA . This study validates another direct target of RNAIII that plays a role in virulence . It also illustrates the diversity of RNAIII-mRNA topologies and how these multiple RNAIII-mRNA interactions would mediate virulence regulation .
Staphylococcus aureus is ubiquitous in the environment and is a commensal organism found on human skin . This major human pathogen is the most common cause of hospital- and community-acquired infections . Therefore , S . aureus has developed a plethora of strategies to survive in various environmental niches . The broad range of human infections caused by S . aureus is in part due to the production of a large number of virulence factors . These factors mediate cell and tissue adhesion , contribute to tissue damage and spreading , and protect the bacteria against the host immune defense system . Coordinated virulence gene expression is thought to be critical for infection and is orchestrated by multiple factors involving two-component systems , global regulatory proteins , and the quorum-sensing system [1] , [2] . Quorum-sensing regulation in staphylococci is mainly driven by the agr system , which was shown to exert a variety of functions in bacterial physiology and pathogenesis [2] , [3] . The agr system is composed of two divergent transcription units , RNAII and RNAIII . RNAII contains a density-sensing cassette ( agrD and B ) and a two-component sensory transduction system ( agrA and C ) . Upon a threshold level of cell density , the response regulatory protein , AgrA , activates the transcription of its own operon and of the regulatory RNAIII [4] . Recent data show the existence of two distinct agr regulatory circuits; one is RNAIII-independent and the other is RNAIII-dependent [3] . Although RNAIII controls the expression of many virulence factors , the expression of several enzymes involved in carbohydrate and amino acid metabolisms are downregulated by an unknown mechanism that is independent of RNAIII . Furthermore , AgrA directly activates the synthesis of several phenol-soluble modulin ( PSM ) peptides at the transcriptional level [3] . Hence , AgrA and RNAIII act in concert to regulate the synthesis of many proteins in response to cell density , interconnecting metabolism , and virulence gene expression [3] , [5] , [6] . RNAIII has a dual function because it acts as a mRNA that encodes a PSM peptide , Δ-hemolysin , and temporally controls the switch between early expression of surface proteins and late expression of several exotoxins [1] . RNAIII belongs to the class of trans-acting RNAs , which regulate several mRNAs at the post-transcriptional level [7] , [8] . The 5′ domain of RNAIII activates translation of hla mRNA ( encoding α-hemolysin ) by preventing the formation of an intramolecular mRNA structure that sequesters the hla ribosome binding site [1] , [9] . The 3′ end and the central domain of RNAIII ( Fig . S1 ) repress the synthesis of early expressed cell surface virulence factors ( protein A , fibrinogen-binding protein ) as well as the transcriptional regulator , Rot , the repressor of toxins [1] , [10]–[12] . We have previously shown that the 3′ domain of RNAIII , which is the most highly conserved domain , could also form base pairings with coa mRNA encoding staphylocoagulase [12] . Staphylocoagulase is an extracellular protein produced by almost all clinical isolates of S . aureus , which specifically forms a complex with prothrombin , the so-called staphylothrombin , to promote fibrin formation in human plasma . Like the major cell surface protein A , the synthesis of staphylocoagulase is growth-phase dependent , and the protein is produced during exponential growth and rapidly repressed by the agr system [13] . We show here that RNAIII is responsible for the in vivo repression of staphylocoagulase at the post-transcriptional level . This results from a direct interaction of two distant domains of RNAIII with coa mRNA . The complex is formed rapidly and is stable enough to prevent the binding of the ribosomal 30S subunit and , in addition , provides binding sites for the endoribonuclease III . Thus , coa mRNA belongs to the RNAIII-dependent repressed mRNAs that are regulated by a similar mechanism . This work and previous data also illustrate the variety of RNAIII-mRNA topologies that are sufficient to block the access of the ribosome at the initiation step .
Sequence complementarity between RNAIII [nucleotides ( nts ) 391 to 437] and coa mRNA ( nucleotides 15 to 52 ) suggested that the 3′ domain of RNAIII can repress coa expression at the post-transcriptional level through the formation of RNAIII-mRNA interactions [12] . To validate the in vivo relevance of such a mechanism , we analyzed the expression of gene reporter constructs in various S . aureus strains that expressed the wild type RNAIII or truncated versions of RNAIII . The 5′ start of coa mRNA was determined by 5′ rapid amplification of cDNA ends ( RACE ) showing that the 5′ untranslated region contains 35 nucleotides upstream the AUG initiation codon . The entire leader regulatory region of the coa gene , including 88 nucleotides of the coding sequence , was cloned in-frame with the lacZ gene into the pTCV-lac shuttle vector [14] . This construct is under control of an agr-independent promoter ( PrpoB ) . The β-galactosidase activity was determined in the S . aureus strain LUG1467 ( wt , rnaIII+ ) , which express RNAIII and in LUG1457 ( ΔrnaIII ) , which carries a deletion of the rnaIII gene ( Fig . 1A ) . We also measured the synthesis of the β-galactosidase from the coa-lacZ fusion in the strains lacking the rnc gene , encoding the endoribonuclease III ( RNase III ) ( LUG1446 , Δrnc ) , or of hfq gene , encoding the Sm-like Hfq protein ( LUG1445 , Δhfq ) . The β-galactosidase activity was reduced six-fold in the LUG1467 strain ( wt , rnaIII+ ) compared to the LUG1457 ( ΔrnaIII ) strain . Furthermore , Hfq had no significant effect on the RNAIII-dependent repression , while the deletion of rnc alleviated the repression of the coa-lacZ fusion ( Fig . 1A ) . Experiments were also carried out on the LUG1457 strain ( ΔrnaIII ) , complemented with different sets of plasmid pLUG274 expressing either the wild-type RNAIII , the 3′ end domain comprising nts 391 to 516 , RNAIII-Δ13 ( deletion of hairpin 13 ) or RNAIII-Δ7-9 ( deletion of hairpins 7 to 9 ) ( Fig . 1A ) . A control experiment was carried out with the plasmid pE194 with no insert . Unexpectedly , this plasmid slightly decreased the β-galactosidase levels compared to the LUG1457 ( ΔrnaIII ) strain . One explanation would be that the copy number of the pTCV-lac derivative was affected by the presence of the multicopy plasmid , pE194 , even though both plasmids are compatible . However , derivatives of plasmid pE194 producing high levels of wild-type RNAIII , the 3′ domain , or RNAIII-Δ7-9 reproducibly decreased the synthesis of β-galactosidase ( Fig . 1A ) . Conversely , the expression of RNAIII-Δ13 , which lacks the base-pairing complementarities with coa mRNA , did not alter β-galactosidase synthesis ( Fig . 1A ) . We also analyzed the steady-state level of coa mRNA in different S . aureus strains in late-exponential phase ( Fig . S2 ) . The mRNA was not detected in RN6390 ( wt , rnaIII+ ) while its level was significantly enhanced in the isogenic strain lacking rnaIII gene ( ΔrnaIII ) . Of interest , in the Δrnc strain , the level of coa mRNA was reproducibly found to be slightly higher than in the parental wt strain ( Fig . S2 ) . This result suggests that the RNase III-dependent degradation of the mRNA contributes in part to the disappearance of the mRNA pool . Complementation assays were also done with the mutant ΔrnaIII strain transformed with plasmids expressing several variants of RNAIII . The expression of the 3′ domain of RNAIII strongly reduced the level of coa mRNA while significant levels of the mRNA were still detected in the strain expressing RNAIII-Δ13 ( Fig . S2 ) . Taken together , these results strongly suggest that RNAIII and RNase III coordinately repress coa expression at the post-transcriptional level and that the hairpin 13 of RNAIII is essential for the repression . The predicted base-pairing between RNAIII and coa mRNA and the in vivo experiments suggested that the RNAIII-dependent repression of coa mRNA was governed by direct RNAIII-mRNA pairing . We thus mapped the regions of interactions using enzymatic and chemical probing . The conformation of coa mRNA was probed using RNase T1 ( specific for unpaired guanines ) , RNase V1 ( specific for helical regions ) , and several base-specific chemicals such as dimethylsufate ( methylates N1A≫N3C ) , a carbodiimide derivative ( modifies N3U≫N1G ) , and diethylpyrocarbonate ( carboxyethylates N7A ) . Several experiments of RNA structure probing are shown in Fig . 2 . The secondary structure model of coa mRNA , which explains most of the probing data , is comprised of three stem-loop structures connected by unpaired residues ( Fig . 3A ) . The AU-rich hairpin I is of weak stability but is proposed to occur based on the enzymatic cleavage pattern . However , the coexistence of alternative structures in the region encompassing nucleotides 10 to 70 may explain the concomitant presence of RNase V1 cleavages and the reactivity of many nucleotides at one of their Watson-Crick positions . In contrast , the long hairpin structure III located in the coding region of coa mRNA is well supported by the enzymatic cleavage patterns and the non reactivity of the Watson-Crick position of A77 to U85 towards chemicals ( Fig . 2B-E ) . Binding of RNAIII induced changes in the region encompassing the ribosome binding site ( RBS , nucleotides U10 to A48 ) . RNAIII protected the guanines of the Shine-Dalgarno ( SD ) sequence against RNase T1 as well as the nucleotides U10 to U18 and A40 to A48 against chemical modifications ( Fig . 2D , E ) . Concomitantly , RNAIII binding induced new RNase V1 cleavages at positions 39–41 and enhanced reactivity of A21 , A24 , A25 , A29 to A31 at position N1 , of A21 at position N7 , and of U26 and U27 at position N3 in coa mRNA ( Fig . 2D–E , Fig . 3B ) . These reactivity changes in the RBS of coa mRNA most likely resulted from the binding of the hairpin 13 of RNAIII because its deletion in RNAIII conferred no additional effect on the accessibility of the RBS of coa mRNA ( Fig . 2C ) . Binding of coa mRNA to the 3′ domain or to RNAIII induced correlated changes in hairpin 13 . Strong protections were observed at G441 against RNase T1 and at positions 411–415 and 448–449 against RNase V1 ( Fig . 2A ) . Concomitantly , increased RNase V1 cleavages were observed at positions 433–434 and 444 . All these data are thus consistent with the formation of a RNAIII-mRNA duplex that sequestered the RBS of coa mRNA ( Fig . 3 ) . This imperfect duplex involves two consecutive regions of 13 base-pairings interrupted by an internal loop and a bulged adenine 21 ( Fig . 3B ) . Unexpectedly , we also found a second RNAIII binding site restricted to the apical loop III of coa mRNA ( Fig . 3B ) . Binding of RNAIII reduced considerably the RNase T1 cuts at G94-97 and the modifications of the nucleotides UGGGAU98 mediated by the chemicals ( Fig . 2B , E ) . Concomitantly , RNAIII binding induced several RNase V1 cuts at positions 96 to 98 ( Fig . 2B ) . These changes were abolished if the complex was formed between coa mRNA and the RNAIII deleted of hairpins 7 to 9 ( Fig . 2C ) . Furthermore , coa mRNA binding to RNAIII reduced significantly the reactivity of the nucleotides CCCA243 towards DMS in the apical loop 7 of RNAIII . These data are strengthened by the sequence complementarities between the apical loop III of coa mRNA and the hairpin loop 7 of RNAIII and support the existence of a loop-loop interaction ( Fig . 3A , B ) . This interaction is , however , strongly dependent on the formation of the imperfect duplex because the reactivity changes in the hairpin loop III of coa mRNA were significantly decreased if complex formation was performed with RNAIII-Δ13 ( Fig . 2C ) . Molecular modeling of the RNA interaction between the two loops shows an almost continuous stacking from the 3′ side of the helix III of coa mRNA , through the loop-loop intermolecular helix to the helix of the hairpin 7 of RNAIII . The two connecting loops of three and two nucleotides bridge the grooves of the newly formed helix ( Fig . 3C ) . Altogether , the data show that the mRNA-RNAIII complex is composed of a bipartite site , which implies the formation of an imperfect duplex and a loop-loop interaction . The contribution of the two binding sites toward complex formation was further evaluated by gel shift assays . Each experiment has been reproduced four times . In vitro labeled coa mRNA was first incubated with increasing concentrations of RNAIII or its variants ( RNAIII-Δ13 , RNAIII-Δ7–9 , and the 3′ domain ) at 37°C for 15 min ( Fig . 4A ) . This experiment shows that coa mRNA binds to RNAIII or its 3′ domain with a Kd value of around 10 nM . The deletion of hairpins 7 to 9 in RNAIII had only a two-fold effect on the dissociation constant ( around 25 nM ) , while the deletion of hairpin 13 in RNAIII increased significantly the Kd value by one order of magnitude ( around 150 nM ) . The initial rate of wild type RNAIII binding to 5′ end-labeled coa mRNA was estimated from a time-course analysis and resulted in an association rate constant of 1 . 1×105 M−1 s−1 ( Fig . 4B ) . Similar values were observed for three other RNAIII-mRNA ( spa , SA1000 , rot ) target complexes [11] , [12] . These data indicate that the complexes are rapidly formed as observed for several fully complementary antisense-target RNA systems [15] , [16] . We then investigated whether deletion of hairpin 7 or hairpin 13 of RNAIII involved in the binding would affect binding rates ( Fig . 4B ) . The binding rate constant for the mutant RNAIII-Δ7-9-coa mRNA pair was identical to the wild type complex ( 9 . 35×104 M−1 s−1 ) . However , for the mutant RNAIII-Δ13-coa mRNA pair , the value was significantly decreased by one order of magnitude lower ( 1 . 1×104 M−1 s−1 ) . These experiments strongly suggest that initial pairings involved the hairpin 13 of RNAIII and that this motif confers stable binding to coa mRNA . Since RNAIII binds to the SD sequence of coa mRNA , we analyzed whether RNAIII binding is sufficient to prevent the formation of the ternary initiation complex formed with the S . aureus 30S subunit , initiator tRNAMet , and coa mRNA . Formation of the ternary complex , which blocked the elongation of a cDNA primer by reverse transcriptase , produced a toeprint at U49/A50 , 15 nucleotides downstream of the initiation codon ( [17]; Fig . 5A ) . Intriguingly , a second toeprint resulting from ribosome binding was also observed at A14; this weak toeprint was not detected with the E . coli ribosomal 30S subunit ( result not shown ) . Binding of RNAIII , RNAIII-Δ7–9 ( deleted of hairpins 7 to 9 ) , or RNAIII-Δ14 ( deleted of hairpin 14 ) strongly decreased the two toeprint signals . This indicates that the regulatory RNAIII totally blocks access of the ribosome at the RBS site of coa mRNA . The inhibition was observed whether the RNAIII-mRNA complex was pre-formed or RNAIII was added together with the 30S subunit ( Fig . 5A ) . This shows that the resulting inhibitory complex is rapidly formed and sufficiently stable to prevent the formation of the ribosomal initiation complex . Using this assay , we were not able to analyze the contribution of the loop-loop interaction in the inhibition of ribosome binding because the primer used for elongation hybridized in the long hairpin loop III . However , the RNAIII-ΔH13 exerted no inhibitory effect on ribosome binding , showing that the specific RNAIII-mediated inhibition of ribosome binding to coa mRNA resulted mainly from the sequestration of the RBS by the hairpin 13 of RNAIII ( Fig . 5A ) . As RNase III is required for efficient repression in vivo ( Fig . 1 ) , we analyzed whether this enzyme can cleave the complex in vitro . We have shown previously that cleavage assays by RNase III can be a useful tool for probing in vitro RNA-RNA complexes [18] . The RNase III-dependent cleavages were probed on the 5′-end labeled RNAs as well as on the native RNAIII-coa mRNA complex using a purified His-tagged RNase III from S . aureus ( Fig . 5B–C ) . Only weak RNase III cleavages were observed in the free coa mRNA . When the 5′ end-labeled mRNA was incubated with RNAIII , four major cleavages occurred at positions 32 , 34 , 39 , and 97 in the mRNA ( Fig . 5B–C ) . Binding of coa mRNA induced a RNase III-dependent cleavage at C241 of the labeled RNAIII ( results not shown ) . Thus , the two regions of hybridization were susceptible to RNase III cleavages . Using truncated versions of RNAIII and the isolated hairpin 7 or the 3′ domain , we were able to assign the partners involved in the RNAIII-mRNA complex . Indeed , the hairpin 7 only induced a specific RNase III-cleavage at position 97 of the mRNA , while the hairpin 13 binding promoted major cleavages at positions 32 , 34 , and 39 of the mRNA ( Fig . 5C ) . Furthermore , the complex formed between RNAIII-Δ7–9 and coa mRNA was cleaved efficiently by RNase III at positions 32 , 34 , and 39 of the mRNA ( Fig . 5B ) . Conversely , only one RNase III-mediated cleavage was detected at position 97 of coa mRNA bound to RNAIII-Δ13 or to hairpin 7 ( Fig . 5B–C ) . This cleavage was , however , weaker than the cleavage found in the wild type complex . These experiments correlate well with the probing data showing that the loop-loop interaction is stabilized by the duplex formed between the RBS of coa mRNA and the hairpin 13 of RNAIII . In the irregular duplex , RNase III cleaves only from the mRNA side , whereas the enzyme induces cleavages on both strands of the loop-loop interaction leading to the classical two nucleotides 3′ overhang . Taken together , these data fully support the chemical and enzymatic probing showing that the hairpin 13 of RNAIII binds to the RBS of coa mRNA , while the hairpin loop 7 forms limited base pairings with the coding sequence . The data further indicate that the loop-loop interaction adopts a topology that is appropriate for efficient RNase III binding and catalysis [18] .
S . aureus produces a large variety of virulence factors that are required for the successful colonization of the host and that confer to the bacteria the ability to counteract the immune defense system of the host [2] . Among these virulence factors , staphylocoagulase primarily activates prothrombin , inducing the formation of a fibrin clot around the bacterial cell [19] . Coating the bacteria with host proteins contributes to hiding the bacteria from the immune system and from phagocytosis . The expression of coagulase was shown to follow a temporal regulation , as do several adhesins and surface proteins that are expressed earlier than the secreted enzymes , immunotoxins , and cytotoxins [2] . Furthermore , coagulase belongs to the early expressed virulence factors such as protein A , the fibrinogen-binding protein SA1000 , and the SsaA-like protein SA2353 , which were found to be repressed by the quorum sensing-controlled RNAIII . During the growth cycle , the level of RNAIII varies inversely with that of coa mRNA [13] . In addition , it was shown that the coagulase expression was both positively and negatively controlled by an agr- dependent mechanism . A functional agr element resulted in a relative elevation of the coa mRNA level at the early exponential phase of growth followed by a strong decrease of the mRNA level at the post-exponential phase of growth [13] . We demonstrate here that the agr-dependent repression effect on coa mRNA is most probably the result of a direct binding of RNAIII to coa mRNA . We show that RNAIII in conjunction with RNase III are required to fully repress the synthesis of staphylocoagulase at the stationary phase of growth ( Fig . 6 ) . The primary effect of RNAIII would be to prevent translation initiation subsequently followed by the RNase III-dependent cleavage of the repressed mRNA . Since we have previously shown that RNase III binds efficiently to RNAIII , we propose that RNAIII-dependent translation repression and RNase III cleavage are coupled . Hence , these data , together with previous works , show that RNAIII represses the synthesis of coagulase , protein A , SA1000 , SA2353 , and Rot by a similar mechanism [10]–[12] . In addition , probing the mRNA structure also shows that coa mRNA adopts a very similar structural organization to spa and SA1000 mRNAs ( Fig . S3 ) . The three mRNAs have short 5′ untranslated regions , which carry a 5′ hairpin structure with a strong SD sequence located in the apical loop ( Fig . S3 ) . In the absence of RNAIII , these elements may confer to the mRNAs a high stability [11] . Indeed , in B . subtilis , stabilization of mRNAs was shown to be a consequence of the blocking of the 5′ end by a stalled initiating ribosome at a SD-like sequence [20] , [21] or by a stable 5′ hairpin structure and a strong RBS [22] . Therefore , the coordinated action of RNAIII and RNase III would be needed to irreversibly repress the synthesis of these virulence factors at an appropriate time . In vitro binding assays show that RNAIII binds to coa mRNA and its other mRNA targets with a rather high association rate constant . Efficient repression by non-coding RNAs ( ncRNA ) , which act at the translational level , requires that the ncRNA binds to target mRNAs within a short time frame , i . e . before the formation of the stable ribosomal initiation complex [15] , [16] . Our data also indicate that S . aureus Sm-like Hfq protein is not required for the RNAIII-dependent repression of coa mRNA in vivo ( Fig . 1 ) , in contrast to Escherichia coli and Salmonella typhimurium ncRNAs which act in concert with Hfq to bind mRNA targets [23] , [24] . Despite the fact that Hfq binds to RNAIII [11] , the observation that the deletion of hfq does not exhibit severe phenotypic defects rules out the direct involvement of Hfq in regard to RNAIII-mediated regulation in S . aureus [10] , [25] . Instead , we propose that the structures of RNAIII and its mRNA targets may compensate for the need of a helper protein as shown for antisense RNAs fully complementary to their target mRNAs [15] , [16] . We , however , do not rule out that another protein or RNase III could contribute to stabilize and/or facilitate the formation of the hybrid [11] . The regions of interaction in RNAIII and coa mRNA contained stem-loop structures that are indeed well appropriate for initial loop-loop interactions . The two conserved C-rich loops , 7 and 13 , of RNAIII bind to the RBS and to the hairpin loop III in the coding sequence of coa mRNA , respectively ( Fig . 3 , 6 ) . These C-rich hairpin loops of RNAIII are also used to repress the other mRNA targets , although the topologies of the resulting inhibitory complexes are different ( Fig . S3 ) . RNAIII forms long duplexes with the RBS of spa and SA1000 mRNAs , while it forms two loop-loop interactions with the 5′UTR and the RBS of rot mRNA , respectively ( [11] , [12] , Fig . S3 ) . Here we show that the RNAIII-coa mRNA complex involves an imperfect duplex of two stretches of 13 base pairs separated by a bulged loop that sequestered the RBS , and a loop-loop interaction that took place in the coding region . In contrast to rot mRNA , in which the two loop-loop interactions were essential for in vivo repression , the sequestration of the RBS of coa mRNA is sufficient by itself to promote efficient repression in vivo and to prevent the formation of the ribosomal initiation complex ( Fig . 5A , 6 ) . Indeed , the loop-loop interaction is not essential for efficient in vivo repression and contributes only moderately to the stability of the inhibitory complex . Hence , the various topologies of the repressed RNAIII-mRNA complexes depend largely on the mRNA context . The inhibitory RNAIII-coa mRNA complex also provided specific binding sites for the double strand-specific RNase III , which induced strong cleavages in the two regions of coa mRNA bound to RNAIII . Notably , the cleavage sites in the loop-loop interaction also occurred at a similar position in the two kissing interactions that took place in the rot mRNA-RNAIII complex ( Fig . S3; [12] ) . The sequences of coa mRNA involved in the loop-loop interactions are very similar to rot mRNA , showing that similar signatures exist in various RNAIII-repressed mRNAs ( Fig . S3 ) . Molecular modeling of the kissing interaction , which took into account the chemical and enzymatic probing data , revealed that the loop-loop interaction induces a coaxial stacking of the two intramolecular helices ( Fig . 3C ) . The overall topology is very similar to the RNA loop-loop structure obtained by NMR , which mimics the interaction between sense and antisense RNAs involved in the regulation of the ColE1 plasmid [26] . Such a long helical structure might well be appropriate for the binding of the homodimeric enzyme , although the sequence of the kissing interactions might also be a specific binding determinant . Of interest , coa mRNA was shown to be completely depleted as soon as RNAIII was produced , and the deletion of rnc caused the accumulation of coa mRNA ( [13]; Fig . S2 ) . Therefore , as we postulated previously , RNase III might initiate rapid degradation of coa mRNA , and the cleavage in the loop-loop interaction may also contribute to access to several other endo- or exoribonucleases for further degradation ( Fig . 6 ) . Notably , at a similar position , spa and SA1000 mRNAs carry a long stem-loop structure in the coding sequence that is also cleaved efficiently by RNase III [11] , [12] ( Fig . S3 ) . In addition , the depletion of the mRNA might also result from an indirect effect of RNAIII . Indeed , Rot protein was shown to activate the transcription of coa mRNA [27] , while RNAIII represses the synthesis of Rot at the post-transcriptional level [10] . Thus , the RNAIII-mediated repression of coagulase would occur at both transcriptional and post-transcriptional levels as it was shown for spa mRNA [10] , [12] , [27] . It is not an exception that RNAIII utilizes conserved C-rich loops to target similar regions of various mRNAs that are functionally related . In S . typhimurium , GcvB RNA represses translation initiation of multiple target mRNAs by binding to a C/A-rich motif present in all these mRNAs , which encode periplasmic substrate-binding proteins of ABC uptake systems for amino acids and peptides [28] , [29] . E . coli CyaR contains a hairpin loop with a conserved anti-SD sequence that is used to target the SD sequence of a subset of mRNAs [30] , [31] . Similarly to RNAIII , we recently found that other S . aureus ncRNAs carry a similar UCCC signature always present in an unpaired region , and through its unpaired C-rich motif , one of these RNAs binds to the RBS and represses the expression of several mRNAs [32] . S . aureus coa mRNA and the other mRNA targets of RNAIII carry a strong SD sequence located in an unpaired region that is quite appropriate for the docking of the 30S subunit , but also for the formation of initial contacts with the C-rich loop of RNAIII ( Fig . S1 ) . Specificity for coa regulation is mainly given by the propagation of the intermolecular contacts to form a long imperfect duplex further stabilized by a loop-loop interaction in the coding sequence . In conclusion , this study validates another direct target of RNAIII that plays a role in virulence . Our study further stresses that the RNAIII harbors highly conserved regions that provide a specific signature to generate interactions with the RBS of multiple mRNAs and that the mRNA context directs the topology of the inhibitory complexes . Recent works focusing on E . coli and S . typhimurium show that regulatory RNAs that target mRNAs regulate gene expression through a variety of unusual mechanisms and bind to mRNA regions located far away from the ribosome binding site in the 5′UTR [33] , in the coding sequence [34] , [35] , and in the 3′ end [36] . Whether S . aureus has also evolved such a diversity of RNA-dependent regulatory mechanisms remains to be addressed .
S . aureus RN6390 or LUG1467 derives from 8325-4 . In WA400 and LUG1457 ( ΔrnaIII ) , the P3 operon is deleted and replaced by the chloramphenicol transacetylase gene ( cat86 ) [37] . LUG774 and LUG911 strains derive from RN6390 , in which rnc and hfq genes , respectively , have been replaced by cat86 gene [11] . Staphylococci were grown either on BM agar plates ( 1% peptone , 0 . 5% yeast extract , 0 . 1% glucose , 0 . 5% NaCl , 0 . 1% K2HPO4 ) or in brain-heart infusion ( BHI ) with erythromycin ( 5 µg/ml ) when appropriate . RNAIII and its variants were expressed in Staphylococcus aureus WA400 with plasmid pE194 ( see Table 1 ) . Translation fusions were constructed with plasmid pLUG220 , a derivative of pTCV-lac , a low-copy-number promoter-less lacZ vector ( Table 1 ) . The 5′ end of the coa mRNA was first determined by rapid amplification of cDNA ends ( RACE ) using the First Choice RLM-RACE kit following the company's protocol ( Ambion ) . The whole leader region of coa mRNA including 126 nt of the coding sequence , was cloned downstream the rpoB promoter in frame with lacZ [11] . β-galactosidase activity was measured three times on duplicate cultures with the Enzyme Assay System ( Promega ) . Electrophoresis of total RNA ( 20 µg ) was done on a 1% agarose gel containing 2 . 2 M formaldehyde and vacuum transfer to nylon membrane . Hybridizations with specific digoxigenin-labeled RNA probes complementary to coa mRNA and luminescent detection were carried out as described previously [13] . RNAIII , RNAIII derivatives ( RNAIII-Δ7–9: deletion of nts G207 to U319 , RNAIII-Δ13: deletion of nts U409 to A451 , and RNAIII-Δ14: deletion of nts G483 to C511 , the 3′ domain comprises nts 391 to 516 ) , the isolated hairpin 7 , and the coa mRNA fragment were transcribed in vitro using T7 RNA polymerase as described previously [38] . The transcribed RNAs were purified by 8% polyacrylamide-8 M urea gel electrophoresis . After elution in 0 . 5 M ammonium acetate/1 mM EDTA buffer , the RNAs were precipitated twice with ethanol . Before use , the pellet was dissolved in sterile bi-distillated water and the concentration was measured accurately . The 5′ end-labeling of dephosphorylated RNA or DNA oligonucleotides was performed with T4 polynucleotide kinase and [γ-32P]ATP [39] . Before use , RNAs were renatured by incubation at 90°C for 2 min in the absence of magnesium and salt , 1 min on ice , followed by an incubation step at 20°C for 15 min in TMN buffer ( 20 mM Tris-acetate pH 7 . 5 , 10 mM magnesium-acetate , 150 mM Na-acetate ) . Binding rate constant of RNAIII-coa mRNA complex was measured as described previously [40] . Binding of end-labeled coa mRNA to a ten-fold excess of unlabeled RNA ( RNAIII , RNAIII-Δ13 , RNAIII-Δ7-9 ) was performed at 37°C in TMN buffer . Samples were withdrawn at various time points ( 0–10 min ) , added to gel application buffer and loaded onto a native 5% polyacrylamide gel . The gel was run at 4°C and constant voltage ( 300 V ) for 3 h and subsequently dried . Bands corresponding to the RNAIII-coa mRNA complex and free RNAIII , respectively , were quantified using the SAFA algorithm [41] . For determination of the dissociation rate constant of RNAIII-coa mRNA complex , end-labeled coa mRNA was incubated with an increased molar amount of wild-type RNAIII or RNAIII variants ( RNAIII-Δ13 , RNAIII-Δ7-9 , 3′ domain , hairpin 7 ) for 15 min at 37°C in TMN buffer . Samples were then treated as described above . All experiments were done four times giving reproducible data . RNAIII-coa mRNA formation was carried out at 37°C for 15 min in TMN buffer . Enzymatic hydrolysis was performed in 10 µl of TMN , in the presence of 1 µg carrier tRNA at 37°C for 5 min: RNase T1 ( 0 . 0025 units ) , RNase V1 ( 0 . 5 units ) . Chemical modifications were performed on 2 pmol of coa mRNA or RNAIII at 20°C in 20 µl of reaction buffer containing 2 µg of carrier tRNA . Alkylation of C ( N3 ) and A ( N1 ) positions was done with 1 µl DMS ( diluted 1/8 and 1/16 in ethanol ) for 2 min , and modification of A ( N7 ) was done with 4 µl of DEPC for 20 min at 20°C in TMN buffer . Modifications of U ( N3 ) and G ( N1 ) were done with 5 µl of CMCT ( 50 mg/ml ) for 10 and 20 min in a buffer containing 50 mM Na-borate pH 8 , 5 mM MgAc , and 150 mM KOAc . RNase III purification and the enzymatic cleavage assays on coa mRNA and on RNAIII were performed as described previously [18] . End-labeled RNA fragments were sized on 12% polyacrylamide/8 M urea slab gels . Cleavage positions were identified using RNase T1 and alkaline ladders of the probed RNA . The cleavage or modification sites of unlabeled RNAs were detected by primer extension . Details for hybridization conditions , primer extension , and analysis of the data have been described previously [38] . S . aureus 30S subunits were prepared according to [42] . The formation of a simplified translational initiation complex with mRNA and the extension inhibition conditions were strictly identical to those described by [38] , [42] . Standard conditions contained 15 nM coa mRNA annealed to a 5′ end-labeled oligonucleotide complementary to nts 99 to 117 of coa mRNA , 250 nM S . aureus 30S ribosomal subunits ( 250 nM ) , and 25 to 100 nM of RNAIII or its variants in 10 µl of buffer containing 20 mM Tris-acetate , PH 7 . 5 , 60 mM NH4Cl , 10 mM magnesium acetate , and 3 mM β-mercaptoethanol . After 10 min at 37°C , the initiator tRNA ( 1 µM ) was added and the reaction was incubated for a further 5 min at 37°C . Reverse transcription was conducted with one unit of AMV reverse transcriptase for 15 min at 37°C . Relative toeprinting ( toeprint band over full-length RNA+toeprint ) was calculated by scanning of the gel with the Bio-imager Analyser ( Fuji ) . Modeling of the regions encompassing residues U73 to A114 of the coa mRNA and of residues A223 to U256 of RNAIII were carried out as described [43] , [44] . Following the interactive assembly step , several cycles of geometrical least-square refinements were performed until a satisfactory solution was reached . Figure 3C was prepared using the PYMOL program ( DeLano WL , The PyMOL Molecular Graphics System 2002; http://www . pymol . org ) . | Staphylococcus aureus causes a wide spectrum of diseases in humans and is one of the main causes of community- as well as hospital-acquired infections . S . aureus produces a large number of virulence factors that are expressed in a coordinated manner and at appropriate time and space . To this end , a set of multiple trans-acting regulators , including regulatory proteins and RNA , is brought into play . The ability of organisms to use RNA to modulate gene expression is a relatively new concept . This is the case for the largest regulatory RNA , S . aureus RNAIII , which controls the switch between the expression of surface proteins and excreted toxins . Here we used a combination of approaches in vivo and in vitro to analyze the mechanism used by RNAIII to regulate the expression of one major virulence factor , staphylocoagulase , which promotes clotting of human plasma . We found that RNAIII regulates the expression of staphylocoagulase through direct interactions with its mRNA . RNAIII binds to two distant regions of coa mRNA to arrest translation and in a coordinated manner , the endoribonuclease III recognizes the formed duplex to initiate degradation of the repressed mRNA . Staphylocoagulase belongs to the early expressed virulence factors , such as protein A , that are repressed by RNAIII using a similar dual mechanism . This study illustrates the diversity of RNAIII-mRNA topologies and how these multiple RNAIII-mRNA interactions would mediate virulence regulation . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"molecular",
"biology/post-translational",
"regulation",
"of",
"gene",
"expression",
"molecular",
"biology/mrna",
"stability",
"molecular",
"biology/translational",
"regulation",
"microbiology",
"molecular",
"biology"
] | 2010 | Staphylococcus aureus RNAIII Binds to Two Distant Regions of coa mRNA to Arrest Translation and Promote mRNA Degradation |
Stress is a fundamental aspect of aging , as accumulated damage from a lifetime of stress can limit lifespan and protective responses to stress can extend lifespan . In this study , we identify a conserved Caenorhabditis elegans GATA transcription factor , egl-27 , that is involved in several stress responses and aging . We found that overexpression of egl-27 extends the lifespan of wild-type animals . Furthermore , egl-27 is required for the pro-longevity effects from impaired insulin/IGF-1 like signaling ( IIS ) , as reduced egl-27 activity fully suppresses the longevity of worms that are mutant for the IIS receptor , daf-2 . egl-27 expression is inhibited by daf-2 and activated by pro-longevity factors daf-16/FOXO and elt-3/GATA , suggesting that egl-27 acts at the intersection of IIS and GATA pathways to extend lifespan . Consistent with its role in IIS signaling , we found that egl-27 is involved in stress response pathways . egl-27 expression is induced in the presence of multiple stresses , its targets are significantly enriched for many types of stress genes , and altering levels of egl-27 itself affects survival to heat and oxidative stress . Finally , we found that egl-27 expression increases between young and old animals , suggesting that increased levels of egl-27 in aged animals may act to promote stress resistance . These results identify egl-27 as a novel factor that links stress and aging pathways .
Responses to various forms of stress play an important role in aging and longevity . Several types of stress result in damage that can accumulate over time ( e . g . oxidative stress results in damaged proteins that often accumulate with age ) [1]–[3] . Responses to these stresses have protective effects that can alleviate the effects of damage accumulation . Consistent with this idea , previous studies have found that mutants with extended longevity often exhibit increased stress resistance [4]–[7] . For example , mutations that disrupt the insulin/IGF-1 like signaling ( IIS ) pathway not only extend longevity , but also increase resistance to many types of stress including heat , oxidative , and pathogenic stress [8]–[11] . As the first genetic pathway in Caenorhabditis elegans that was linked to longevity , the IIS pathway is a conserved endocrine component that controls important aspects of development , metabolism , and stress response [12] . Activation of the IIS receptor ( DAF-2 ) causes phosphorylation of a phosphatidylinositol 3-kinase ( AGE-1 ) , which initiates a cascade of signals resulting in phosphorylation and inactivation of a FOXO transcription factor ( DAF-16 ) . Reduction of IIS through knockdown of daf-2 or through the presence of certain environmental stresses , results in activation of DAF-16/FOXO , which triggers a transcriptional program that promotes both stress resistance and longevity [10] , [12] . Many genes in the DAF-16 transcriptional response are involved in various stress responses , and some of these also change in expression during aging . For example , heat shock proteins are induced by many types of stress including heat and pathogenic infection [13]–[17] , and expression levels of certain heat shock genes are increased in C . elegans mutants with reduced IIS and extended longevity . Furthermore , heat shock proteins in C . elegans increase in expression between young and old animals , although expression is reduced in very old populations in which 90% of the population is dead [18] , [19] . The increased expression of stress genes during aging is not confined to worms . Studies have shown that genes induced by oxidative stress increase with age in flies [20]; p53-related damage response genes increase with age in mice [21]; also , genes that are involved in immunological complement activation , which are generally induced in response to oxidative and pathogenic stress [22]–[26] , increase expression in old age across four human tissues [27] . While these results suggest that stress response pathways become increasingly activated in old organisms , it is unclear whether this activation has a protective function and is beneficial for longevity or whether it represents a misregulation of stress pathways and is a contributor to organismal decline . In C . elegans , the upstream regions of the genes that constitute the DAF-16 transcriptional program are enriched for both the DAF-16 binding site and a GATA-like transcription factor binding site [28] . One of the GATA factors that may be involved in the DAF-16 mediated IIS transcriptional program is ELT-3 , as elt-3 expression is increased in age-1 mutants . Furthermore , elt-3 is required for the longevity phenotype of daf-2 mutants , suggesting that the elt-3/GATA transcription factor functions downstream of the IIS pathway [29] . The GATA family of transcription factors may also play important roles in regulating the molecular changes that accompany normal aging . Transcriptional profiling of young and old animals has revealed that the promoters of age-dependent genes are enriched for GATA motifs . The GATA transcription factor elt-3 is responsible for some of the age-dependent changes in gene expression . Expression of elt-3 declines as worms age , resulting in decreased expression of its downstream targets . Low levels of elt-3 have a deleterious effect on survival and stress response suggesting that this decline in elt-3 levels hastens the aging process [29] . In this work , we identify another GATA transcription factor , egl-27 , that functions to promote stress survival and to delay aging . In addition to its homology to GATA factors , egl-27 is also homologous to the MTA1 component of NuRD chromatin remodeling complex [30]–[33] . Previous studies show that egl-27 is expressed in most somatic cells during development and in adult worms [30] , [34] . We show that egl-27 expression increases with age and that increased levels of egl-27 through overexpression are sufficient to extend lifespan and to increase survival in response to heat stress . In contrast , reducing egl-27 activity suppresses the longevity and thermotolerance phenotypes of reduced insulin/IGF-1 like signaling . Moreover , egl-27 can respond to the presence of stress as its expression is induced by a variety of different stresses . EGL-27 binds upstream of genes involved in both stress and aging , but interestingly EGL-27 targets are enriched for genes whose expression decreases with age . Finally , egl-27 expression is regulated by the GATA transcription factor elt-3 and the IIS transducing gene daf-16 . These results define egl-27 as a novel factor that promotes both longevity and stress response .
Reduction of egl-27 activity by RNAi knockdown was previously shown to partially suppress the longevity phenotype of the long-lived IIS receptor mutant daf-2 ( e1370 ) [29] . We extended this result by showing that the cold-sensitive egl-27 ( we3 ) allele could fully suppress the longevity phenotype of daf-2 ( e1370 ) . Specifically , we found that daf-2 ( e1370 ) ; egl-27 ( we3 ) double mutants have a median lifespan that is 2 . 5 fold shorter than daf-2 ( e1370 ) single mutants and 35% shorter than wild-type worms ( Figure 1A ) . egl-27 ( we3 ) mutants live only 10% shorter than wild-type worms , suggesting that the combination of the egl-27 ( we3 ) mutation and the daf-2 ( e1370 ) mutation results in a slight synthetic lethality ( Figure 1A ) . As a control , we compared the extent of daf-2 ( e1370 ) suppression by egl-27 ( we3 ) to that of daf-16/FOXO , a well-characterized suppressor of daf-2 [35] . We found that daf-2 ( e1370 ) ; daf-16 ( mu86 ) double mutants have a median lifespan that is two-fold shorter than daf-2 ( e1370 ) single mutants and 19% shorter than wild-type worms ( Figure 1A ) . These data show that egl-27 ( we3 ) suppresses daf-2 ( e1370 ) longevity to approximately the same extent as daf-16 ( mu86 ) . egl-27 ( we3 ) is cold-sensitive for lethality [30] , and so we tested whether temperature affects the suppression of daf-2 longevity by egl-27 ( we3 ) . We hatched worms at the developmentally permissive temperature ( 20°C ) and then shifted them at day 2 of adulthood to 15°C , 20°C , or 25°C . We found that egl-27 ( we3 ) suppresses daf-2 ( e1370 ) longevity at all three temperatures; specifically , daf-2 ( e1370 ) ; egl-27 ( we3 ) worms have a median lifespan that is 2 . 9 fold shorter than daf-2 ( e1370 ) mutants at 15°C , 1 . 9 fold shorter at 20°C , and 1 . 7 fold shorter at 25°C ( Figure S1A ) . These results show that egl-27 ( we3 ) is temperature-sensitive for developmental arrest but not for suppression of longevity by daf-2 ( e1370 ) . We next tested whether increased levels of egl-27 are sufficient to increase longevity . To do this , we engineered three different constructs containing egl-27 and generated strains overexpressing each construct . The first construct is from the modENCODE project and contains GFP-tagged egl-27 in a fosmid with 18 kb of sequence upstream and 8 kb of sequence downstream of egl-27 . This fosmid also contains the full coding sequence for three other genes: F31E8 . 6 , F31E8 . 1 , and tbc-1 . We found that worms expressing egl-27::GFP had a lifespan extension of 13% ( Figure 1B ) . The second construct contains mCherry-tagged egl-27 with full intergenic regions covering 5 kb of sequence upstream and 152 bp of sequence downstream of egl-27 . Worms overexpressing egl-27::mCherry had a lifespan extension of 15% ( Figure 1B ) . Finally , we cloned the egl-27 genomic region containing the we3 temperature-sensitive mutation from egl-27 ( we3 ) worms . This construct also contains full intergenic regions . We generated three transgenic worm strains containing egl-27 ( we3 ) on an extrachromosomal array in order to create strains that conditionally overexpress egl-27 at the permissive temperature . We grew worms at either the non-permissive or permissive temperature starting at day two of adulthood , and then measured their lifespans . Interestingly , we found that overexpression of egl-27 ( we3 ) extended lifespan at both the permissive and non-permissive temperatures . Specifically , at 20°C , median lifespan was increased 23–31% ( Figure S1B ) and at 15°C , the median lifespan was increased 11–21% ( Figure S1C ) . These results suggest that the addition of low levels of egl-27 ( we3 ) activity at 15°C is sufficient to extend lifespan or that egl-27 ( we3 ) is temperature sensitive for development but not for its life-extending functions . To determine whether egl-27 is expressed at higher levels in these overexpression lines compared to control worms , we used qRT-PCR to measure levels of egl-27 mRNA expression in the different overexpression strains and in control worms . We found that egl-27 expression is increased 2 . 4 fold in the egl-27::GFP strain versus control worms ( p = 0 . 0008 , Figure S5E ) . egl-27 expression is increased 4 . 1 fold in the egl-27::mCherry strain ( p = 0 . 02 , Figure S1D ) . Finally , egl-27 expression is increased 23% , 11% , and 2 . 4 fold in the three egl-27 ( we3 ) overexpression strains although none of these increases are significant , possibly due to high expression variability caused by extrachromosomal array expression ( Figure S1D ) . We did not observe any abnormal developmental phenotypes in any of the egl-27 overexpression lines , suggesting that these levels of increased egl-27 do not adversely affect development . Furthermore , we validated that transgenic egl-27::GFP can function like endogenous egl-27 ( + ) , as we showed that egl-27::GFP can rescue the egl-27 ( we3 ) lethal mutant phenotype ( Figure 1C ) . Increased longevity is strongly correlated with increased stress resistance [4] , [36] . To determine whether egl-27 can promote stress survival , we assessed the relationship between egl-27 activity and survival to heat and oxidative stress . To assay the phenotype of an egl-27 reduction-of-function mutation , we used the hypomorphic mutation egl-27 ( we3 ) . To assay the phenotype from overexpression of egl-27 , we used the egl-27::GFP strain described above . We assayed heat-stress survival by subjecting worms to 8 hours at 35°C and then measuring survival ( Figure 2A ) . egl-27 reduction-of-function mutants die more quickly after heat stress than wild-type worms; egl-27 ( we3 ) has a median survival time following heat-stress that is 2 . 9 fold shorter than for wild-type worms ( log rank p-value = 7 . 8×10−8 ) . egl-27 gain-of-function worms survive longer after heat stress than control worms; they have a median survival time following heat stress that is 1 . 4 fold longer than for control worms ( p = 5 . 8×10−6 ) and they also have a time to 95% mortality that is 2 . 6 fold longer than for control worms . daf-2 insulin-like receptor mutants are resistant to many types of stress [9] , [11] . We showed that egl-27 ( we3 ) partially suppresses the heat resistance phenotype of daf-2 . daf-2 ( e1370 ) ; egl-27 ( we3 ) double mutants have a median survival that is 1 . 7 fold shorter than daf-2 ( e1370 ) single mutants , but still 7 . 5 fold longer than wild-type worms . In contrast , a well-characterized suppressor of daf-2 , the FOXO transcription factor daf-16 ( mu86 ) fully suppresses the heat stress resistance conferred by daf-2 mutants . daf-2 ( e1370 ) ; daf-16 ( mu86 ) double mutants have a median survival that is 12 . 9 fold shorter than daf-2 ( e1370 ) single mutants and that is the same as wild-type worms ( Figure 2B ) . Additionally , egl-27 ( we3 ) has a less pronounced effect on the heat stress survival of daf-2 ( e1370 ) mutants than it does on wild-type worms ( 1 . 7 fold shorter vs . 2 . 9 fold shorter respectively ) suggesting that egl-27 may be required for some but not all of the heat-stress resistance of daf-2 ( e1370 ) mutants . We examined whether egl-27 has a functional role in mediating the response to oxidative stress . To do this , we grew worms on plates supplemented with 10 mM paraquat and then measured their survival time . We observed that altering levels of egl-27 did not have a large effect on oxidative stress survival as neither reduction nor gain of egl-27 activity affects median lifespan compared to control worms ( Figure 2C ) . However , we found that egl-27 ( we3 ) partially suppresses the oxidative stress resistant phenotype of daf-2 ( e1370 ) ; specifically , daf-2 ( e1370 ) ; egl-27 ( we3 ) double mutants have a median lifespan that is 2 . 5-fold shorter than for daf-2 ( e1370 ) mutants and that is 1 . 4-fold longer than wild-type worms ( Figure 2D ) . Although the median survival time for daf-2 ( e1370 ) ; egl-27 ( we3 ) double mutants is similar to that of daf-2 ( e1370 ) ; daf-16 ( mu86 ) double mutants ( which have a median time of survival that is 2 . 3-fold shorter than for daf-2 ( e1370 ) mutants and that is 1 . 5-fold longer than for wild-type worms ) , the time to 95% mortality is different between the two lines . While daf-2 ( e1370 ) ; daf-16 ( mu86 ) double mutants have a time to 95% mortality that is 4 . 1 fold shorter than for daf-2 ( e1370 ) single mutants , daf-2 ( e1370 ) ; egl-27 ( we3 ) double mutants have a time to 95% mortality that is 16% shorter than for daf-2 ( e1370 ) single mutants ( Figure 2D ) . These results suggest that egl-27 activity is important for some of the oxidative stress resistance conferred by reduced activity of the insulin-like receptor gene daf-2 . The results presented above suggest that egl-27 functions downstream of daf-2 in the IIS pathway . To test whether the IIS pathway can modulate egl-27 expression , we examined whether egl-27 expression is altered in daf-2 mutants . Using fluorescence microscopy , we compared the intestinal expression of an integrated egl-27::mCherry transcriptional reporter in a strain with reduced daf-2 activity to control worms in two day old adult , hermaphrodite worms ( Figure 3A ) . We found that egl-27 expression increased 58% in daf-2 ( e1370 ) mutants compared to control worms , providing molecular evidence that egl-27 is regulated by daf-2 in the IIS pathway ( Figure 3B ) . To further define where egl-27 acts in the IIS pathway , we examined whether egl-27 acts downstream or upstream of the IIS pathway modulator daf-16/FOXO . To do this , we tested whether a mutation in daf-16 suppresses the increased levels of egl-27 found in daf-2 mutants ( Figure 3A ) . We found that levels of egl-27::mCherry in daf-2 ( e1370 ) ; daf-16 ( mu86 ) double mutants are 2 . 1 fold lower than in daf-2 mutants ( Figure 3B ) . This indicates that daf-16 is required for increased egl-27 expression in daf-2 mutants , which suggests that egl-27/GATA is regulated by daf-16/FOXO in the insulin signaling pathway . Further supporting this idea , we found a canonical DAF-16/FOXO binding element ( GTAAACA ) [28] , [37] 614 bps upstream of the egl-27 translational start site , suggesting that egl-27 may be a direct target of DAF-16 . In addition to daf-16/FOXO , the GATA transcription factor elt-3 modulates IIS , as elt-3 expression is increased in age-1 mutants with reduced IIS and elt-3 is partially required for the longevity phenotype of daf-2 mutants [29] . To determine whether egl-27 acts downstream of the GATA transcription factor elt-3 , we tested whether egl-27 expression is affected by a null mutation in elt-3 ( Figure 3A ) . We found that levels of egl-27::mCherry are 3 . 0 fold lower in elt-3 ( vp1 ) mutants compared to control worms , suggesting that egl-27 acts downstream of elt-3 ( Figure 3B ) . We next tested whether elt-3 is required for increased levels of egl-27 expression found in daf-2 mutants . We found that levels of egl-27::mCherry in daf-2 ( e1370 ) ; elt-3 ( vp1 ) double mutants are 6 . 7 fold lower than in daf-2 mutants and 4 . 8 fold lower than control worms ( Figure 3B ) . Furthermore , levels of egl-27::mCherry in daf-2 ( e1370 ) ; elt-3 ( vp1 ) double mutants are similar to levels in elt-3 ( vp1 ) mutants ( Figure 3B ) , suggesting that elt-3 is necessary for heightened egl-27 expression in the context of reduced IIS . In the expression experiments above , we measured egl-27::mCherry expression in the anterior intestine . To determine whether egl-27 is regulated in a tissue-specific manner , we also examined egl-27 expression in the head region , which is composed of several different cell types – hypodermal , neuronal , muscle , and pharyngeal cells . We found that egl-27::mCherry levels are 35% higher in daf-2 ( e1370 ) mutants and 34% lower in elt-3 ( vp1 ) mutants compared to control worms . egl-27::mCherry levels are 37% lower in daf-2 ( e1370 ) ; daf-16 ( mu86 ) double mutants and 68% lower in daf-2 ( e1370 ) ; elt-3 ( vp1 ) double mutants compared to daf-2 ( e1370 ) worms ( Figure S2A ) . These results suggest that IIS and elt-3 regulate egl-27 expression across multiple tissues , although the magnitude of this regulation may vary slightly across tissues . We also determined whether egl-27 regulation by IIS and elt-3 GATA transcription occurs only during adulthood , or whether the same regulation occurs during development . To do this , we examined whether egl-27 expression is affected by mutations in daf-2 and elt-3 at the L2 larval stage of development . We found that egl-27 levels are 80% higher in daf-2 ( e1370 ) mutants and 3 . 1 fold lower in elt-3 ( vp1 ) mutants compared to control worms ( Figure S2B ) . Because egl-27 is regulated by daf-2 and elt-3 to approximately the same degree during development and adulthood , the genetic networks that regulate egl-27 expression are likely developmental programs that persist into adulthood . To confirm our fluorescent microscopy results , we also examined how endogenous egl-27 levels are affected in daf-2 and elt-3 mutants . Using qRT-PCR , we showed that egl-27 levels are 67% higher in daf-2 ( e1370 ) mutants ( p = 0 . 009 ) and 2 . 0 fold lower in elt-3 ( vp1 ) mutants ( p = 0 . 003 ) compared to control worms during the L2 larval stage of development ( Figure S2C , S2D ) . To determine whether egl-27 forms a feedback loop with daf-16 and elt-3 , we examined whether egl-27 can act upstream of these regulators . To examine whether egl-27 can regulate either DAF-16 or ELT-3 activity , we examined how reduction of egl-27 affects levels of expression of sod-3::GFP , an established transcriptional reporter for DAF-16 [38] , [39] and ELT-3 [29] , [39] ( Figure 3C ) . We found that levels of a sod-3::GFP transcriptional reporter are not significantly different between egl-27 ( we3 ) mutant worms and control worms ( Figure 3D ) . However , DAF-16 activity is low in wild-type worms , so we examined whether reduction of egl-27 affects levels of sod-3::GFP expression in daf-2 mutants where DAF-16 is highly activated . Similar to previous reports [38] , [39] , we found using fluorescent microscopy that sod-3 expression is 4 . 3 fold higher in daf-2 ( e1370 ) worms compared to control worms , and that increased sod-3 expression is suppressed in daf-2 ( e1370 ) ; daf-16 ( mu86 ) double mutants ( Figure S2E ) . We found that sod-3 levels are equally high in daf-2 ( e1370 ) ; egl-27 ( we3 ) double mutants ( Figure 3D ) . These results indicate that reduction of egl-27 activity does not affect expression of sod-3::GFP , suggesting that neither DAF-16 nor ELT-3 activity are affected in egl-27 ( we3 ) mutants . Finally , we examined whether egl-27 can regulate its own expression . To do this , we examined how reduction of egl-27 activity affects the expression of an egl-27::mCherry transcriptional reporter ( Figure 3E ) . We found that egl-27::mCherry expression is reduced by 32% in egl-27 ( we3 ) mutants compared to control worms ( Figure 3F ) . This suggests that egl-27 activates its own expression in a feed-forward loop . Supporting this , we found several GATA-like binding motifs in the promoter region of egl-27 ( TTATC/GATAA 107 bps upstream , TATCA/TGATA 728 bps upstream , and CTTATCA/TGATAAG 800 bps upstream of the translational start site ) . These results suggest a role for GATA transcription factors such as ELT-3 or EGL-27 itself , in directly regulating egl-27 expression . In response to many types of stress , DAF-16/FOXO becomes activated [40] , [41] . This leads to increased protection from the stress itself , and may also lead to increased levels of egl-27 expression . According to this model , various types of stresses could also lead to changes in expression of egl-27 , via activation of DAF-16/FOXO . To test this possibility , we exposed worms carrying an egl-27::mCherry transcriptional reporter to six different stresses ( osmotic shock , gamma radiation , starvation , heat stress , oxidative damage , and UV damage ) . We then compared egl-27::mCherry expression under each stress condition to its expression in controls using fluorescent microscopy ( Figure 4A ) . We found that egl-27::mCherry expression is induced after exposure to starvation , heat stress , oxidative damage , and UV stress ( p value<0 . 001 ) ( Figure 4B ) . Interestingly , none of the stresses increase egl-27 expression by more than two-fold . This suggests that egl-27 acts differently from canonical stress-induced genes , such as heat shock proteins , which are expressed at low levels under normal conditions but can be induced up to 100-fold following heat stress [42] , [43] . As a control , we showed that the increase in egl-27 expression is specific and not caused by a general increase in transcription in response to these stresses . We used fluorescence microscopy to measure expression levels of myo-3::GFP , a muscle-specific myosin gene ( Figure S3A ) , and showed that its expression is not induced in response to starvation , heat stress , oxidative damage , or UV stress ( Figure S3B ) . To determine how egl-27 expression changes during aging , we used qRT-PCR to measure egl-27 expression in day 4 and day 14 adult worms . We found that egl-27 expression increases two-fold between young and old worms ( Figure 4C ) . This is consistent with previous work showing that stress genes such as heat shock proteins increase in expression as worms age before declining in expression in extremely old worms [18] , [44] . These data suggest that egl-27 expression , like the expression of other stress-related genes [18] , [20] , [21] , [27] , [45] , increases in old worms . To better understand the mechanism by which egl-27 promotes longevity and stress survival , we identified where EGL-27 binds in the genome . To do this , we prepared lysates of the egl-27::GFP worm strain described above , at the L2 larval stage of development . These lysates were used by the modENCODE consortium to generate binding site data for EGL-27 , using a GFP antibody to immunoprecipate the GFP-tagged EGL-27 . We chose to perform ChIP-seq using L2 larval stage worms rather than adult worms because the majority of datasets generated by modENCODE were for larval stages . Because we showed that egl-27::GFP can rescue the egl-27 ( we3 ) lethal mutant phenotype ( Figure 1C ) , the sites that are bound by EGL-27::GFP are likely to be representative of the sites that are bound by endogenous EGL-27 . By examining ChIP-seq data from the modENCODE project , we identified 4113 DNA regions showing significant binding by EGL-27::GFP [46] . Previous work has shown that some DNA regions are bound by one or a few transcription factors ( factor-specific ) while other DNA sites are associated with a large number of transcription factors , and that the specific functions of each transcription factor are better defined by its factor-specific targets than by these redundantly bound targets [46] . Because we were interested in the factor-specific functions of EGL-27 , we removed 2306 sites located within redundantly bound regions from further analysis . Of the factor-specific binding sites , 481 are located within gene promoters , defined as 1000 bps upstream to 500 bps downstream of a translational start site . 426 binding sites are located within exons and 466 binding sites are located within introns . 78 are located within 1000 bp downstream from the translational stop site . Finally , 516 are located in intergenic regions . Because we were interested in putative targets of EGL-27 , we focused on the 481 peaks located within gene promoters . To identify the consensus sites that may be directly bound by EGL-27 , we examined the 481 ChIP-seq peaks that fall within gene promoters for the presence of enriched DNA motifs . We used the Gibbs sampling program BioProspector [47] to perform a de novo motif search on the center 100 bp sequence of EGL-27 ChIP-seq identified binding sites . The top 10 motifs found by the program are variations of two motifs: the GATA motif and a novel RGRMGRWG motif ( Table S2 ) . The GATA motif ( GAKAAG ) is found in 32% of EGL-27 target peaks and the novel RGRMGRWG motif is found in 25% of target peaks . Both are significant when compared to background sets consisting of randomly generated 100 bp sequences centered from 1000 bp upstream to 500 bp downstream of translational start sites ( Figure 5 ) . Both motifs are also significant when compared to scrambled sequence derived from EGL-27 peaks that preserve nucleotide frequencies ( Figure 5 ) . The novel RGRMGRWG motif is not a consensus-binding site for any known class of transcription factors . However , this motif was previously identified as enriched in the promoters of differentially-expressed genes in insulin signaling daf-2 mutants as well as sirtuin pathway mutants [48] . Since EGL-27 contains a GATA DNA-binding domain , the enrichment for GATA motifs in the EGL-27 binding sites supports its function as a GATA transcription factor . Previous studies have shown that GATA motifs are important for osmotic and pathogenic stress response [49]–[51] and aging [28] , [29] , which is consistent with our model that egl-27 regulates stress and aging genes by binding the GATA motif . The 481 EGL-27 promoter peaks are located in the promoter regions of 501 unique genes ( Table S3 ) . We conducted gene ontology ( GO ) analysis [52] , [53] on the set of 501 EGL-27 target genes in order to determine if the EGL-27 target genes are enriched for specific biological pathways . We found that EGL-27 targets are enriched for hypodermal genes that form the cuticle as well as intestinal genes involved in aromatic compound catabolic process ( Table S4 ) . We examined whether target genes bound by EGL-27 significantly overlap genes that change expression with age . We obtained a list of 1132 age-regulated genes as previously defined by DNA microarrays [29] . Specifically , we computed the hypergeometric p-value for the overlap between the set of 501 EGL-27 target genes and the set of 1132 age-regulated genes . We found that 67 EGL-27 targets show age-dependent expression , which is a 2 . 8 fold enrichment over the number expected by chance ( p = 9 . 5×10−15 ) ( Table 1 ) . Even though the ChIP-seq analysis was performed at the L2 larval stage of development , we were still able to find a strong enrichment for age-regulated genes . Interestingly , 61 of the 67 targets decline in expression with age ( Figure S4 ) suggesting that increased levels of egl-27 during normal aging are insufficient to prevent the age-dependent decline of these genes . Using GO analysis , we found that these 67 age-dependent EGL-27 targets are even further enriched for hypodermal genes that form the cuticle as well as several metabolic categories including aromatic amino acid family metabolic process , oxoacid metabolic process , and organic acid metabolic process ( Table S4 ) . Because we previously showed that egl-27 expression is induced in response to several types of stress , we examined whether egl-27 targets are differentially-expressed in response to different types of stress . To do this , we acquired published transcriptional responses to 16 different stress conditions involving response to six pathogens , six environmental stresses , and four toxins ( Table S5 ) . We then compared each set of genes that are differentially-expressed in that particular stress condition to the set of EGL-27 targets to determine whether the two sets significantly overlap by a hypergeometric test . Even though the ChIP-seq experiment for EGL-27 was not performed under stress conditions , we found that EGL-27 target genes are significantly ( p<10−5 ) enriched for differentially-expressed genes from 11 of the 16 different stress conditions ( Table 1 ) . This supports the idea that egl-27 is involved in the response to many types of stress . To better characterize how egl-27 mediates heat-stress survival , we assessed how egl-27 regulates heat-stress target genes . To do this , we examined the expression of four egl-27 target genes ( grd-3 , T14B1 . 1 , Y37A1B . 5 , and lpr-3 ) that were previously shown by DNA microarray experiments to be differentially expressed following heat stress [54] . We first wanted to see that expression of these genes change as expected after heat stress in wild-type worms . To do this , we extracted RNA from wild-type second larval stage worms that were exposed to 90 minutes of heat at 35°C followed by a 2-hour recovery . We then used qRT-PCR to determine how the expression of each gene changes in heat stressed worms compared to unstressed worms . Similar to the DNA microarray data , we found that grd-3 , T14B1 . 1 , and Y37A1B . 5 expression increases while lpr-3 expression decreases after heat stress ( Figure 6A–6D ) . We examined how reduced egl-27 activity affects target gene expression before heat stress . To do this , we compared expression levels of EGL-27 target genes in egl-27 ( we3 ) mutants to their expression levels in wild-type worms . We found that two genes ( grd-3 and T14B1 . 1 ) had higher expression while two genes ( Y37A1B . 5 and lpr-3 ) had lower expression in egl-27 loss-of-function worms than in wild-type worms ( Figure 6A–6D ) . This shows that egl-27 plays a role in modulating target gene expression under normal conditions when worms are unstressed . Next , we assessed how reduction of egl-27 activity affects the heat stress response of these four genes , where the heat stress response is defined as the ratio of expression after heat stress compared to before stress . We found that reduction of egl-27 activity resulted in attenuation of the heat-induced changes of all four genes ( Figure 6E ) . Interestingly , the reduced heat stress response is not caused by lower levels of induced expression , but rather by higher levels of basal expression . Expression of all four target genes was significantly altered in egl-27 reduction-of-function mutants prior to heat stress ( Figure 6A–6D ) . For example , in egl-27 reduction-of-function worms , the expression of two targets ( grd-3 and T14B1 . 1 ) is induced even in pre-stressed animals . Although expression of these genes remains unchanged following stress , both pre-stress and post-stress expression levels in egl-27 reduction-of-function worms are similar to their levels in post-stress wild-type worms . In the case of these two genes , egl-27 may function to alter baseline expression levels in unstressed worms . However , expression levels of two other targets ( Y37A1B . 5 and lpr-3 ) are significantly different in egl-27 reduction-of-function worms compared to wild-type worms both pre- and post- heat stress . For these two genes , egl-27 may function to alter baseline expression levels in unstressed worms and is required for additional activation of expression following heat stress . These results suggest that endogenous egl-27 is required to reduce basal stress levels in wild-type worms . To complement the reduction-of-function results , we examined how increased basal egl-27 activity in the egl-27::GFP overexpression strain affects differential expression of egl-27 in response to heat-stress ( Figure S5 ) . First , we used qRT-PCR to measure the combined RNA levels of egl-27::GFP and endogenous egl-27 . As expected , the egl-27::GFP overexpression strain had both increased levels of combined egl-27 expression before and after stress when compared to control worms ( Figure S5E ) . Additionally , we observed that the heat stress response of egl-27 is greater in egl-27::GFP worms compared to control worms ( 2 . 9 fold versus 2 . 2 fold respectively ) ( Figure S5E ) . This result suggests that worms with egl-27::GFP have increased levels of egl-27 activity , which may make the expression of egl-27 more sensitive to heat stress as part of a positive feedback loop . However , the egl-27::GFP strain showed variable effects on the expression of EGL-27 target genes both before and after heat stress ( Figure S5A–S5D ) . Expression of grd-3 was unaffected by increased levels of egl-27 before heat stress . Following heat stress , expression of grd-3 was higher in the egl-27::GFP strain than in the control strain , such that induction levels were 1 . 8 fold in the egl-27::GFP strain compared to 1 . 2 fold in the control strain . Expression of T14B1 . 1 was unaffected by increased levels of egl-27 both before and after heat stress . Expression of Y37A1B . 5 was increased in egl-27 gain-of-function worms before heat stress but repressed in these worms following heat stress . Finally , expression of lpr-3 was 19-fold lower in egl-27 gain-of-function worms before heat-stress compared to control worms and 3 . 6-fold lower following heat stress . Whereas heat stress causes this gene to decrease expression slightly in control worms , heat stress causes its expression to increase in the egl-27::GFP strain . While expression of lpr-3 was significantly down-regulated following heat stress in wild-type worms , lpr-3 levels were not significantly changed in transgenic control worms , underscoring the background differences that can exist between wild-type and transgenic animals . These data suggest that increased levels of egl-27 affect the expression of some target genes both prior to and after heat stress .
We have shown that a GATA/MTA1 transcription factor [29]–[33] , [55] , egl-27 , is an important mediator of stress response and longevity in C . elegans . Using binding site data from chromatin immunoprecipitation followed by ultra high throughput sequencing ( ChIP-seq ) , we have shown that EGL-27 binds upstream of genes that are enriched for those that increase with age and that change in response to diverse stresses . We examined whether EGL-27 has a beneficial or detrimental effect on longevity and stress response . We found that overexpression of egl-27 increases both longevity as well as survival in response to heat stress . In contrast , egl-27 mutants have a shortened lifespan and reduced survival in response to heat stress . Furthermore , reduction of egl-27 activity suppresses the longevity phenotype as well as the heat and oxidative stress resistance phenotypes of daf-2 mutants . These results suggest that egl-27 may promote longevity through promoting stress resistance . This possibility is supported by other studies showing that increased expression of genes that confer resistance to specific stresses also extend lifespan . For example , increased levels of zebrafish lysosyme which confers antimicrobial defense [56] , heat shock factor 1 which confers heat resistance [5] , [56] , and skn-1 which confers resistance to oxidative damage [6] , [57] , are all capable of extending lifespan when overexpressed in transgenic worms . Here , we identify a GATA transcription factor , egl-27 , that plays a role in the response to multiple stresses and has a beneficial effect on longevity when overexpressed . Altering levels of egl-27 activity affects the expression of heat stress genes both in unstressed as well as heat-stressed worms . From these experiments , we infer that egl-27 normally maintains a program to resolve cellular stress , and that altering levels of egl-27 alters baseline stress levels in worms . Similar to this , a previous study showed that certain stress response genes are expressed at lower levels in stress-resistant daf-2 mutants [58] . This suggests that that changes that alter baseline levels of stress can also alter baseline expression of stress response genes . egl-27 expression is regulated by a variety of stress pathways . We found that egl-27 expression is induced by multiple stresses: heat stress , oxidative damage , UV irradiation , and starvation . Next , we showed that egl-27 acts downstream of the GATA transcription factor elt-3 and two IIS pathway components ( IIS receptor daf-2 and the FOXO transcription factor daf-16 ) . egl-27 expression is induced in long-lived daf-2 ( e1370 ) mutants and this induction is suppressed in daf-2 ( e1370 ) ; daf-16 ( mu86 ) and daf-2 ( e1370 ) ; elt-3 ( vp1 ) double mutants . These data support our finding that egl-27 ( we3 ) can fully suppress daf-2 ( e1370 ) longevity . Furthermore , previous work has shown that elt-3 is a pro-longevity factor whose expression is confined mainly to hypodermal cells [29] , [59] , [60] . Our finding that elt-3 can regulate egl-27 expression in several tissue types including the intestine suggests that elt-3 can affect gene expression in a cell non-autonomous manner . Surprisingly , expression of sod-3 , which acts as a readout for DAF-16 activity [38] , [39] , is unchanged in daf-2 ( e1370 ) ; egl-27 ( we3 ) double mutants compared to daf-2 ( e1370 ) single mutants , suggesting that activated DAF-16 is not sufficient for extended longevity in the absence of functional egl-27 . Finally , we found that egl-27 can regulate its own expression in a feed-forward loop . This evidence for auto-regulation supports the idea that egl-27 may be involved in a complex circuit with feedback mechanism for regulating target gene expression . Interestingly , egl-27 expression increases with age in wild-type worms . Our finding that increased egl-27 expression extends lifespan and improves stress resistance suggests that the way that egl-27 expression changes during aging is beneficial for the organism . In contrast , previous studies have focused on age-dependent changes in expression that are detrimental for the organism . For example , the GATA transcription factor elt-3 is an important regulator of the transcriptional changes that occur between young and old . elt-3 declines in expression with age and low levels of elt-3 have a deleterious effect on survival and stress response , suggesting that declining levels of elt-3 may act as a driver of aging [29] . Another study found that NF-κB acts as a key regulator of age-dependent gene expression differences in nine types of human and mouse tissues [61] . NF-κB expression increases in old animals , and this increase is detrimental as blocking NK-κB in old skin results in a partial reversal of the aging transcriptome and more youthful skin [61] , [62] . In contrast to elt-3 in C . elegans and NF-κB in mice , our studies suggest that changes in egl-27 expression during aging may act to improve stress response and to promote longevity . However , most EGL-27 binding targets from ChIP-seq experiments decline in expression with age , suggesting that this increased expression is insufficient to prevent the age-dependent decline of these genes . Because increased levels of egl-27 extend lifespan , increased expression of egl-27 in old worms appears to delay the aging process instead of causing it . Our work offers novel insight into the interplay between stress and aging , and suggests that aging is not simply a process of moving from an ideal young transcriptome to an inadequate old transcriptome . Rather , age-dependent changes in gene expression are likely comprised of a mix of beneficial , detrimental , and neutral changes .
All C . elegans strains ( Table S6 ) were handled and maintained as described previously [63] . Genotyping primers are described in Table S7 . Lifespan experiments were conducted as previously described [35] , [64] . All experiments were done at 20°C unless otherwise noted . Age refers to days following adulthood and p-values were calculated using the log-rank ( Mantel-Cox ) method [65] . daf-2 ( e1370 ) and egl-27 ( we3 ) are single base pair change mutants , so we used the tetra-primer ARMS-PCR procedure [66] to design 4 primers for each SNP of interest . daf-16 ( mu86 ) and elt-3 ( vp1 ) are deletions so we used a combination of 3 primers ( two flanking the deleted region and one inside of the deleted region ) to probe for the deletion . To generate DNA for genotyping , 1–10 worms were lysed in 1× PCR buffer with 1 . 5 mM MgCl2 and Protease K . A single PCR reaction was setup using this DNA and all ( three or four ) primers and then visualized on a 2% agarose gel . For egl-27 ( we3 ) , all reactions will produce a 396 bp product; mutant allele will produce a 213 bp product while wild-type allele will produce a 240 bp product . For daf-2 ( e1370 ) , all reactions will produce a 300 bp product; mutant allele will produce 155 bp product while wild-type allele will produce a 203 bp product . For daf-16 ( mu86 ) , mutant allele will produce a 400 bp product while wild-type allele will produce a 634 bp product . For elt-3 ( vp1 ) , mutant allele will produce a 145 bp product while wild-type allele will produce a 401 bp product . Primers and temperatures are detailed in Table S7 . Worms carrying the integrated transgene egl-27::GFP ( OP177 ) were crossed to egl-27 ( we3 ) ( JA1194 ) to generate egl-27::GFP; egl-27 ( we3 ) ( SD1751 ) in the F2 generation , which was identified by PCR genotyping ( genotyping primers for egl-27 ( we3 ) detailed in Table S7; methods for single nucleotide genotyping above ) . N2 , egl-27 ( we3 ) , and egl-27::GFP; egl-27 ( we3 ) worms were synchronized using hypochlorite and grown to day 1 of adulthood . 10 hermaphrodites from each strain were individually placed onto NGM plates seeded with E . coli and allowed to lay eggs for 1 hour . The adult worms were removed and the number of eggs counted . The numbers of hatched worms were scored 1 day , 2 days ( not shown ) , and 6 days ( Figure 1C ) after the beginning of egg laying . % survival was computed as the percentage of hatched worms divided by the number of total eggs . Error bars represent the standard error between the 10 replicates . To quantify mCherry and GFP expression , we imaged at least 15 worms for each condition at 20× using a Zeiss Axioplan microscope . Images were analyzed using ImageJ [67] . All assays were performed using day 1 adult hermaphrodites . In all cases , E . coli refers to the OP50 strain . Control worms were always transferred the same number of times , and imaged at the same time as experimental worms . All strains used for imaging and survival assays are detailed in Table S6 . Two independent cultures of worms expressing egl-27::GFP ( OP177 ) were synchronized using sodium hypochlorite to isolate eggs followed by hatching in S basal overnight . Arrested L1 stage larva were collected and grown on NGM plates seeded with E . coli until mid L2 stage . ChIP-seq was performed by the modENCODE consortium [46] , [69] , [70] . The program PeakSeq [71] was used to identify EGL-27::GFP binding sites ( q value<0 . 00001 ) . Peaks bound by five or more out of the original 23 transcription factors were removed from further analysis . Significant , factor-specific peaks were then compared to the C . elegans genome to identify putative EGL-27 target genes . A gene is identified as a target if the center of a peak occurs 1 . 5 kb upstream or 500 bp downstream of its translational start site . The center 100 bp sequences from the 481 promoter peaks were filtered for low complexity regions using RepeatMasker [72] and then submitted to BioProspector to identify overrepresented cis regulatory sequences [47] . To eliminate motifs or amino acid distributions that are generally enriched in all promoters , a random set containing 481 masked 100 bp promoter sequences was submitted to BioProspector as background input . We report the highest 10 motifs from BioProspector ( Table S2 ) , but since only 2 are unique , we ran further analysis on these two . To determine the fold enrichment and probability of these two motifs occurring in our dataset compared to the probability expected by chance , 1000 datasets containing 481 randomly-generated 100 bp promoter sequences ( to match the number of promoter peaks ) were created . We scored the number of times these two motifs were found in our promoter sequences and the average number of times they were found in each of the 1000 background datasets . An enrichment factor for each motif was calculated using ( # of ChIP-seq peaks containing at least one instance of the motif ) / ( average # of background peaks containing at least one instance of the motif ) . A chi-squared p-value for each motif was calculated using the # of ChIP-seq peaks containing at least one instance of the motif as observed and # of background peaks containing at least one instance of the motif as expected ( Figure 5 ) . As a second control , we generated a set containing the 481 original sequences but scrambled the order of the nucleotides in order to create a random sequence while preserving the nucleotide frequency . We generated 1000 iterations of this set of 481 scrambled sequences and for each motif , scored the average number of times they were found in this scrambled sequences . Again , we calculated an enrichment factor and a p-value for each motif in comparison to the scrambled sequences ( Figure 5 ) . EGL-27 target genes and age-regulated EGL-27 target genes were submitted to GOrilla [52] , [53] for Gene Ontology ( GO ) analysis . A background gene list was also submitted . For EGL-27 target genes , the background list consisted of all annotated C . elegans genes . For age-regulated EGL-27 target genes , the background list consisted of all genes represented on the Stanford C . elegans microarray platform . GOrilla outputs a FDR q-value and an enrichment score for each GO category . The FDR q-value is the hypergeometric p-value corrected for multiple hypotheses testing using the Benjamini and Hochberg method [73] . Enrichment score is computed as ( # of inputted genes in GO category/# of total inputted genes ) / ( # of genes in each GO category/# of background genes ) [52] , [53] . Age-regulated genes were obtained from [29] and probe IDs were remapped to WormBase gene IDs ( WS228 ) using annotations from Stanford Microarray Database . Probes that could not be mapped to gene IDs were removed from further analysis . Differentially-expressed stress response genes for each of 16 stress conditions were obtained from publications detailed in Table S5 . When probe names were supplied , probe IDs were remapped to WormBase gene IDs ( WS228 ) using annotations from Affymetrix , Stanford Microarray Database , and Washington University Genome Sequencing Center . Probes that match multiple genes were assigned all associated gene IDs . Probes that could not be mapped to gene IDs were removed from further analysis . If only gene names were supplied , gene IDs were filtered using WormBase gene annotations ( WS228 ) . Unmatched probe IDs or gene IDs were removed from further analysis . To determine whether EGL-27 target genes were significantly enriched for age-regulated or stress-response gene sets , we first determined the number of EGL-27 targets that are significantly differentially-expressed in each age or stress response set . For each comparison , the background gene count is the number of genes included in the platform for that microarray . Using this number , we computed , both a hypergeometric p-value and an enrichment score for each comparison , where enrichment score is defined as ( # of EGL-27 targets/# of differentially expressed genes ) / ( # EGL-27 targets/# of background genes ) . N2 , egl-27 ( we3 ) , control , and egl-27::GFP ( OP177 ) worms ( full genotypes and strain information in Table S6 ) were synchronized using hypochlorite to isolate eggs followed by hatching in S basal overnight . Arrested L1 stage larva were collected and grown on NGM plates seeded with E . coli until mid-L2 stage before splitting into an experimental and control set . Experimental worms were then incubated at 35°C for 90 minutes and allowed to recover at 20°C for 2 hours before collecting in Trizol ( Invitrogen ) . Control worms were kept at 20°C and collected in Trizol at the same time . Total RNA was isolated using Trizol reagent , treated with DNaseI to degrade genomic DNA , and purified using RNeasy kit ( Qiagen ) . cDNA was synthesized using oligo dT primers and SuperScript II First Strand Kit ( Invitrogen ) . qPCR reactions were performed using RT2SYBR Green qPCR Mastermix ( Qiagen ) and the 7900HT Fast Real-Time PCR System ( ABI ) . Melting curve analysis was performed with each primer pair to ensure that quantification is the result of only one product . A serial dilution was performed for each primer pair to generate a standard curve . act-1 was used as an internal control to normalize expression levels as previously described [74] , [75] . All primers are detailed in Table S7 . | Stress is a fundamental aspect of aging , but it is unclear whether the molecular mechanisms underlying stress response become altered during normal aging and whether these alterations can affect the aging process . In this study , we found a GATA transcription factor called egl-27 , whose targets are significantly enriched for age-dependent genes and stress response genes , and whose expression increases with age . In contrast to previous work describing factors that are causal for aging , we found that egl-27 activity is likely beneficial for survival since egl-27 overexpression extends lifespan . egl-27 promotes longevity by enhancing stress response; specifically , increased levels of egl-27 protect animals against heat stress , while reduced egl-27 activity impairs survival following heat and oxidative stress . These results suggest that aging is not simply a process of constant decline . Some factors , such as egl-27 , are more active in old animals , working to restore organismal function and to improve survival . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"medicine",
"insulin-dependent",
"signal",
"transduction",
"cellular",
"stress",
"responses",
"anatomy",
"and",
"physiology",
"physiological",
"processes",
"developmental",
"biology",
"organism",
"development",
"signaling",
"pathways",
"biology",
"aging",
"signal",
"transduction",
"cell",
"biology",
"physiology",
"genetics",
"molecular",
"cell",
"biology",
"genetics",
"and",
"genomics"
] | 2012 | The GATA Transcription Factor egl-27 Delays Aging by Promoting Stress Resistance in Caenorhabditis elegans |
A wide range of organisms features molecular machines , circadian clocks , which generate endogenous oscillations with ~24 h periodicity and thereby synchronize biological processes to diurnal environmental fluctuations . Recently , it has become clear that plants harbor more complex gene regulatory circuits within the core circadian clocks than other organisms , inspiring a fundamental question: are all these regulatory interactions between clock genes equally crucial for the establishment and maintenance of circadian rhythms ? Our mechanistic simulation for Arabidopsis thaliana demonstrates that at least half of the total regulatory interactions must be present to express the circadian molecular profiles observed in wild-type plants . A set of those essential interactions is called herein a kernel of the circadian system . The kernel structure unbiasedly reveals four interlocked negative feedback loops contributing to circadian rhythms , and three feedback loops among them drive the autonomous oscillation itself . Strikingly , the kernel structure , as well as the whole clock circuitry , is overwhelmingly composed of inhibitory , rather than activating , interactions between genes . We found that this tendency underlies plant circadian molecular profiles which often exhibit sharply-shaped , cuspidate waveforms . Through the generation of these cuspidate profiles , inhibitory interactions may facilitate the global coordination of temporally-distant clock events that are markedly peaked at very specific times of day . Our systematic approach resulting in experimentally-testable predictions provides insights into a design principle of biological clockwork , with implications for synthetic biology .
A variety of living organisms on Earth features built-in molecular clock machineries that control the organism’s daily activities [1] . These internal time-keepers , circadian clocks , generate endogenous oscillations of gene expression with ~24 h periodicity , enabling the anticipation of diurnal environmental variations and the coordination of biological processes to the optimal times of day . Examples of such biological processes include sleep/wake cycles in animals , emergence from the pupal case in fruit flies , spore formation in fungi , and leaf movements in plants [2–4] . Disruption of circadian rhythmicity is associated with a wide range of pathophysiological conditions , indicating the importance of clock functions in homeostasis [5–8] Compared to other organisms , such as fungi , insects , and mammals whose circadian systems have been well studied , a molecular understanding of the plant circadian system is still elusive . Numerous molecular and genetic approaches using Arabidopsis thaliana have facilitated the discovery of more than 20 plant clock genes as well as their regulatory interactions [1 , 9 , 10] . The emerging picture from this effort suggests that the core regulatory circuit of the plant circadian system is more complex than in other organisms [9 , 11–13] . The apparent complexity of the plant clock machinery raises a fundamental question: are all the regulatory interactions between clock genes equally necessary for the establishment and maintenance of plant circadian rhythms ? In other words , can we distinguish more important from less important regulatory interactions for normal clock functioning ? Answering this question involves an attempt to prioritize our focus amongst numerous regulatory interactions , in order to simplify a global view of , and thereby elicit an essential principle of , the plant clock organization . Despite the fundamental importance of this issue , a satisfactorily systematic approach has not been taken yet; thus , this topic is the focus of our study . In the case of other biological processes , finding essential subnetworks out of the whole has been of wide interest for both scientific and engineering purposes [14–18] . Properly designed experiments may be one way to address this issue , but often require laborious and costly efforts . Complementary to experiments , mathematical models help biological findings by predicting the effects of genetic and non-genetic perturbations , where experimental access could be limited or unavailable . Utility of mathematical models has been well documented in earlier studies of circadian rhythms [19–22] . An initial mathematical model of the plant circadian system was constructed based only on three genes , LATE ELONGATED HYPOCOTYL ( LHY ) , CIRCADIAN CLOCK ASSOCIATED 1 ( CCA1 ) , and TIMING OF CAB EXPRESSION 1 ( TOC1 ) [22] . This model has evolved to include five times more components to date [23 , 24] . Additionally , models that incorporate the downstream targets of the core circadian system are starting to gain attention [25] . These models have certainly served a significant role in enhancing our understanding of the plant circadian clock . Nevertheless , to the best of our knowledge , none of these studies has fully attempted to specify the functionally essential interactions between clock genes in a systematic and comprehensive way . Central to our approach to the plant circadian system is the concept of a kernel . We define a kernel as a collection of minimal functional sets , each comprising all molecular components ( genes and gene products ) in the system and only a part of their regulatory interactions , which must be present to generate the temporal trajectory of molecular concentrations close to wild type ( WT ) . In this definition , we refer to a collection of minimal sets to cover cases with multiple minimal sets . Based on an Arabidopsis clock model constructed in this study , our analysis shows that the kernel structure combines four negative feedback loops whose interplay effectively accounts for circadian rhythmicity in Arabidopsis . Strikingly , the kernel structure , as well as the whole clock circuitry , was found to be overwhelmingly composed of inhibitory interactions between genes . We subsequently present a mechanistic reason for the prevalence of such inhibitory interactions in the plant clock . These results provide a systematic and unique view of the plant circadian oscillators , with experimentally testable predictions to enhance our understanding of biological time .
We began by constructing a mathematical model of the core circadian oscillator in plant Arabidopsis thaliana . For this model construction , we applied system identification techniques to publicly available time course data of mRNA and protein expression ( Materials and Methods ) . The resulting model consists of 24 ordinary differential equations ( ODEs ) , describing a rate of a concentration change of each mRNA , protein , or protein complex ( S1 Text ) . Experimentally-verified molecular interactions were primarily incorporated in the model , which then contains a total of 40 transcriptional and post-translational interactions between components , along with light-dependent regulations . Fig 1A shows a global architecture of the core gene circuit considered in our model . In comparison with previous models [23 , 24 , 26] , the new model is mainly based on the model ( P2013 ) by Pokhilko et al . [23] , but we filtered out hypothetical or outdated molecular interactions and adopted some recent findings [24] . Compared to our earlier work [26] , which uses a discrete-time model for control design purposes , here we have constructed a continuous-time model , with revised interactions compatible with recent knowledge . Full details of the model comparisons are presented in S1 Text . Overall , we stress that our current model does not intend to outperform other existing models in its accuracy through the inclusion of all up-to-date information . Rather , the priority was to construct a model which is compact , yet biologically relevant , in accordance with recent experimental knowledge . We expect that this model is suitable enough for our main purpose of kernel identification , without further sophistication of the model structure . Because we are ultimately moving forward to identify the kernel structure responsible for circadian rhythms in WT plants , time series data of mRNA and protein expression from WT , not from mutants , were used during model construction . Mutant data were used only to validate the constructed model , as will be described later . Specifically , we estimated the parameters of the model by fitting the simulation results to WT mRNA and protein expression profiles over time , under five different light conditions: equal length light-dark cycle , i . e . , 12 hours of light and 12 hours of dark ( 12L:12D ) , 16 hours of light and 8 hours of dark ( long day ) , 8 hours of light and 16 hours of dark ( short day ) , constant light ( LL ) , and constant dark ( DD ) . These expression profiles were obtained from publicly available experimental literature and databases ( S1 Table ) . Because the absolute levels of mRNAs and proteins were difficult to ascertain from their sources , we normalized the expression levels into dimensionless values ( ≤1 ) with arbitrary scales . As a proxy for the LHY/CCA1 information , we adopted the LHY expression data , because they were often better in the quality than CCA1’s . Constraining the model output to fit all these datasets gave rise to a total of 97 estimated parameters of the model equations , along with 51 coefficients that scale each light condition’s mRNA and protein levels relative to the levels under 12L:12D cycles ( see S1 Text ) . Our model does not separate nuclear from cytosolic proteins [27 , 28] , due to incomplete availability of the relevant expression data and to avoid increasing model complexity . What is the resulting performance of our model ( MF2015 ) ? We found that MF2015 captures well the overall temporal patterns of gene expression from WT ( Fig 1B–1G; for comparison with P2013 , see S1 Text ) . Also , the free running rhythms in WT are in good agreement with experimental values [29 , 30]: 25 . 2 h ( model ) and 24 . 6 h ( experiment ) in LL , and 25 . 8 h ( model ) and 25 . 9 h ( experiment ) in DD . However , these results cannot validate MF2015 , because we estimated the model parameters from the WT data . To directly test the predictive power of the model against an independent dataset , we computed the altered rhythmicity under different genetic perturbations . The simulated mutants are 76 . 2% accurate when the clock periods are quantitatively compared to experimental values ( see S1 Text ) . Qualitative agreement ( lengthened period , shortened period , or arrhythmia ) is observed for 85 . 7% of the simulation outcomes and experimental results ( S1 Text ) . Moreover , the simulation predicts the substantial elevation ( reduction ) of ZEITLUPE ( ZTL ) protein levels in LL ( DD ) , matching the experimental finding [31] . This result is the first accurate reproduction of ZTL performance through computational modeling ( S6 Fig ) . Taken together , MF2015 is greatly supported by an array of experimental evidence in terms of its predictability . Note that P2013 yields the simulated mutant periods in 42 . 9% quantitative agreement with experimental values . In general , the simulation outcomes were robust to a wide range of kinetic parameter variations and transient molecular concentration changes ( S1 Text ) . A few exceptions that convey the system’s sensitive response involve the variations of parameters in PSEUDO RESPONSE REGULATOR 5 ( PRR5 ) mRNA degradation , EARLY FLOWERING 3 ( ELF3 ) inhibition by LHY/CCA1 , and light-responsive protein production . Whether they represent genuine biological factors or model incompleteness is unknown . Meanwhile , the overall robustness to parameter variations indicates the presence of multiple parameter sets for the model . Interestingly , alternative parameters that we examined did not make much of an improvement in the predictability of mutant period lengths ( S1 Text ) . Moreover , such alternative parameters of the model are unlikely to change the main results of our study , as kernel identification and analysis involve parameter re-optimization processes . Our modeling of the core circadian system ( MF2015 ) encouraged us to address difficult mechanistic questions . Among all 40 molecular interactions and light regulations in the system , which interactions ( and light regulations ) are minimally necessary to shape the circadian mRNA and protein expression profiles observed in WT across different light conditions ? We refer to this collection of minimal sets as the kernel of the circadian system . In the next paragraph , both molecular interactions and light regulations are referred to simply as interactions . Sheer screening of interaction sets , where removal severely distorts clock rhythmicity , would not be sufficient to identify a kernel structure . If this distortion is repaired by a readjustment of kinetic parameters , the removed interactions are not likely to be essential in their network-topological properties; rather , their knockout effect is simply dependent on specific parameters . Therefore , the knockout effect in distorting clock rhythms should be double-checked with re-optimized parameters . If the knockout effect remains severe even after parameter re-optimization , the removed interactions can now be said to be essential in their network topological properties . Ideally , our kernel discovery procedure would be to search through all possible combinations of interactions , and examine the effects when the interactions in each combination are removed , followed by parameter re-optimization to best fit the WT expression profile of every clock component across different light conditions . This strategy , although ideal , is extremely computationally demanding and therefore impractical . Instead , we devised a heuristic approach that consists of the following steps ( Materials and Methods , and S1 Text ) : first , we measure the knockout effect of each interaction on the WT expression patterns under the five different light conditions . Then , we prune those interactions from weak to strong knockout effects until discovering any single clock component that fails to produce rhythms similar to WT . Next , among the remaining interactions , we choose those with knockout effects below a certain threshold . Each chosen interaction is deleted , and parameter re-optimization follows to fit the WT expression data . If parameter re-optimization recovers the WT rhythms for every clock component , this interaction is completely removed from the system . The implementation of these steps , complemented by an additional step to allow multiple solutions , leaves a fraction of the interactions , which yet connect all the molecular components in the system . This interaction set corresponds to our estimated kernel structure . For a detailed description of the kernel identification , see S1 Text . Using MF2015 , we found that the kernel of the plant circadian system consists of 22 transcriptional and post-translational interactions and light regulations , which seamlessly involve all molecular clock components in the system . In other words , at least half of the 40 interactions/regulations in the whole system are required to form the WT rhythms across the five different light conditions . Notably , the kernel structure harbors four negative feedback loops , termed loops I to IV ( Fig 2; compare with Fig 1A ) . In the kernel , the only negative feedback other than these four loops is the autoinhibition of the EVENING COMPLEX ( EC ) genes through the EC , and this effect remains localized to the EC formation and thus not our focus here . Loops I to IV host at least one of the PSEUDO RESPONSE REGULATOR ( PRR ) genes each , and are interlocked by having LHY/CCA1 in common: loop I includes LHY/CCA1 , PRR5 , and TOC1 ( Fig 2A ) . Loop II has LHY/CCA1 , PRR7 , and TOC1 ( Fig 2B ) . Loop III involves LHY/CCA1 , PRR7 , and the EC , along with the EC subcomponents ( Fig 2C ) . Lastly , loop IV includes LHY/CCA1 , and PRR9 regulated by light ( Fig 2D ) . Accordingly , TOC1 interconnects loops I and II , while PRR7 interconnects loops II and III . Each of loops I , II , and III includes a cyclic structure of triple inhibitions , known as a repressilator ( Fig 2A–2C ) [32] . A repressilator structure can exhibit sustained oscillation under proper conditions . Of note , loop I has one more interaction added to this repressilator structure , i . e . , the inhibition of PRR5 by LHY/CCA1 . The direction of this inhibitory interaction is exactly opposite to the repressilator’s overall cyclic direction , and thus is supposed to be antagonistic to the oscillatory capability of the loop ( see below ) . Among the four loops , loop IV in Fig 2D is the simplest one , having only a pair of single positive and negative connections between two morning-expressed components , coupled with light . To our knowledge , loops I and II have not been previously described , whereas loop III recapitulates a repressilator structure previously reported [33] . Loop IV has been previously termed the morning loop [9 , 34 , 35] . Therefore , our unbiased and systematic approach to kernel identification does not only recover previously characterized gene circuits ( loops III and IV ) , but also suggests new circuits ( loops I and II ) that may be crucial for Arabidopsis clock function . Owing to the above kernel identification , the complex plant clock circuitry has been greatly simplified , converging on the four negative feedback loops that structure the kernel . We next considered an in-depth mechanistic analysis of the individual feedback loops as well as their interrelations . An immediate question is , among the four negative feedback loops , which of the loops critically support the generation of autonomous molecular oscillations observed in WT . By definition , every element in the kernel must play a significant role in shaping the oscillatory profiles . However , it does not mean that their contributions to the creation of the autonomous oscillation are necessarily equivalent to each other . Moreover , the current kernel structure is a full repertoire of interactions necessary for all five different light conditions mentioned above . Clearly , only separate simulations of constant , free running conditions will answer this question for the endogenous , autonomous oscillation . To test the capability of individual loops to generate autonomous oscillations close to WT , we simulated LL using a computational model of each isolated loop , with kinetic parameters re-optimized for the WT expression data in LL ( S1 Text ) . Given the WT expression profiles , this parameter re-optimization was expected to reveal the maximum oscillatory capacity of each loop structure regardless of its specific MF2015 parameters . It infers a natural bound of the loop’s contribution to the WT endogenous oscillations − a natural bound imposed by the loop’s structure itself rather than by specific parameters . From this simulation , we found that loops I , II , and III in LL were clearly able to generate sustained oscillations similar to WT ( Fig 3A and 3B ) , whereas loop IV failed ( Fig 3C ) . In fact , if equipped with other parameters , oscillations can be maintained even by loop IV , but at the expense of its specific oscillatory patterns , in far deviation from the experimental profiles . Once loop IV undergoes a parameter adjustment to fit the experimental profiles , it loses sustained oscillation . The endogenous oscillatory capability of individual loops I to III raises an intriguing possibility: can the plant circadian rhythm be robust to the breakage of some loop ( s ) , if buffered by the other loop ( s’ ) activity ? To explicitly address this question , we inactivated loop I in MF2015 by blocking the inhibition of LHY/CCA1 by PRR5 . Likewise , we inactivated both loops II and III simultaneously , by blocking the inhibition of LHY/CCA1 by PRR7 . The MF2015 simulation of LL demonstrates that either of these two “mutations” largely restores the circadian gene expression profiles observed in WT , if accompanied by parameter re-optimization ( Fig 3D ) . As can be predicted , the simultaneous blockage of both PRR5 and PRR7’s inhibitory actions on LHY/CCA1 in MF2015 inactivated all three oscillatory loops I to III , and thus abolished the circadian rhythmicity itself of gene expression , even when accompanied by parameter re-optimization . This prediction is well supported by an experimental report that the Δprr5/prr7 double mutant in constant conditions exhibits almost arrhythmic mRNA levels of clock-controlled genes , although each single mutant retains free running rhythmicity [36] . Moreover , the above simulation forecasts that only the removal of the two inhibitory interactions , rather than the entire double gene deletion , is necessary to cause severely abnormal clock gene expression . In sum , we find that under certain circumstances loop I can buffer the loss of loops II and III , and vice versa . Similarly , we computationally blocked PRR7 inhibition by TOC1 , and that by the EC , to inactivate loop II and loop III , respectively . Again , simulated mutant outcomes suggest that loop II and loop III can buffer the loss of each other . Taken together , these results indicate complementary relationships between loops I , II , and III in the management of endogenous circadian oscillations . While loops I to III exhibit the fundamental capacity to generate endogenous oscillations similar to WT , loop IV lacks such capability . We therefore conjectured that , among all the four loops , loop IV is unlikely to exert the strongest regulation on the clock gene expression , if these genes are regulated by the other loops as well . Indeed , the LHY/CCA1 inhibition by PRR9 ( in loop IV ) was consistently weaker than either the LHY/CCA1 inhibition by PRR7 or that by PRR5 ( in loops I to III ) , throughout our simulation with various re-optimized parameters ( S1 Text ) . Previous experimental data from LL have shown that a Δprr9 knockout has a smaller effect on LHY and CCA1 expression than a Δprr7 knockout [29] . Fig 3E shows that LHY mRNA levels , on average , increased by 60 . 5% and 16 . 7% in the Δprr7 and Δprr9 mutants , respectively , consistent with our computational prediction; a similar trend was also observed for CCA1 mRNA [29] . Despite the loop IV’s relatively weak role in free running rhythmicity , it should be noted that , in our current kernel structure , loop IV is the only negative feedback loop which senses external light stimulus ( Fig 2D ) and thereby contributes to the entrainment of the kernel dynamics to light . We cannot entirely exclude the possibility that more loops may come into play in light sensing of the kernel as our model becomes updated . The efficacy of our simple kernel structure to interpret the clock dynamics is further exemplified by loop I . In addition to the basic repressilator structure , loop I holds a unique topological feature of reciprocal inhibitory interactions between LHY/CCA1 and PRR5 ( Fig 2A ) . In particular , the inhibition of PRR5 by LHY/CCA1 is placed in opposition to the repressilator’s overall cyclic direction , and thus may retard the loop’s inherent oscillation . In fact , this retardation effect was found to affect the oscillation of the whole clock circuitry , because of the structural interconnection between loop I and the whole . For example , the simulation of MF2015 in LL demonstrates that a 20% increase in PRR5 inhibition by LHY/CCA1 slows down the circadian rhythm , resulting in a 3 . 3 h lengthened period , whereas a 20% decrease in this inhibition shortens a period by 2 . 9 h ( Fig 3F ) . This experimentally-testable idea might be hard to conceive without the simplicity of the loop-I structure . In the kernel , LHY/CCA1 interlocks all loops I to IV , indicating its central role in the circadian oscillator . The adverse effect of the Δlhy/cca1 double knockout on model performance is supported by experimental evidence [37 , 38] . From the entire kernel structure in Fig 1A , compared with loops I to III , one can notice the presence of TOC1 inhibition by the EC . This inhibition is the only regulatory interaction with its regulated target ( TOC1 ) in the loops , while the interaction itself is not a part of major negative feedback loops in the kernel . This fact prompted us to investigate whether TOC1 inhibition by the EC should be retained in our kernel . The simulated removal of this inhibition from the kernel apparently distorted , e . g . , the LHY mRNA and TOC1 protein profiles , even when accompanied by parameter re-optimization ( S7 Fig ) . Therefore , we keep in the present kernel structure TOC1 inhibition by the EC . In conclusion , our model is supported by current experimental data and indicates that the plant circadian oscillator is an orchestrated interaction of mainly four negative feedback loops in the kernel . In the face of the larger complexity of the full circuitry , our simplified loop structures may offer an efficient way to understand the plant clock mechanisms , as well as predict circadian dynamics that has not yet been characterized . Among the four major negative feedback loops in the kernel , loops I to III have the repressilator-like structures that are entirely composed of inhibitory interactions . Only loop IV includes an activating interaction . Regarding the central role of these feedback loops in circadian rhythms , why does the plant circadian system favor such inhibitor-enriched loops for its function ? Indeed , recent molecular studies of the plant circadian system have indicated that inhibitory relationships outnumber activating regulations among all clock genes [39] . The full circuitry considered in MF2015 is dominated by inhibitory interactions , and this feature becomes even more prominent in its kernel structure , harboring only one activating interaction ( Fig 1A ) . The dominance of such inhibitory interactions distinguishes the plant clock from other circadian systems , including those of mammals and fungi , which have comparable numbers of inhibitory and activating interactions [11–13] . This issue can begin to be addressed by considering that the kernel structure is designed for the production of temporal gene expression patterns close to WT ( Fig 4A ) . Therefore , we presumed that many inhibitory regulations , at least in the kernel , may generate specific waveforms of the WT expression profiles . We do observe , in fact , that a number of Arabidopsis clock genes often exhibit particular waveforms of mRNA and protein expression ( Figs 1B–1G and S1–S5 ) . This waveform is characterized by an asymmetry between the acrophase and bathyphase , as schematized in Fig 4B: the acrophase shows a relatively sharpened peak , whereas the bathyphase can be approximated as flat . Regarding the overall acuteness around a particular peak phase , we here describe this pattern as cuspidate . For comparison , a common sinusoidal wave is not cuspidate , having a symmetrically rounded shape to the acrophase and bathyphase . To examine the possible relevance of inhibitory regulation in cuspidate waveforms , we created a mathematical system consisting of a single transcription factor , either an inhibitor or activator , and its own target gene ( Fig 4A and Materials and Methods ) . We formulated the model equations similar to MF2015 . On the assumption that the target gene shows a near cuspidate ~24h-period expression pattern of proteins ( Fig 4B ) , we conversely asked what specific abundance profile the transcription factor ( inhibitor or activator ) should have for the production of that target gene profile . Our simulation results highlight a clear difference between inhibitor and activator cases , when the target gene exhibits a cuspidate pattern ( S1 Text ) . The inhibitor or activator tends to have a large or small phase difference , respectively , of ~8 to 12 hours or ≲4 hours with the target gene in their protein profiles , as shown in Figs 4C–4E and S8 . In other words , an inhibitor ( activator ) and its target have a roughly antiphase-like ( inphase-like ) relationship . Otherwise , the target gene’s protein expression waveform will not be cuspidate but will exhibit a more smoothened profile ( S9 Fig ) . These facts were initially observed in our simulation with simplified , yet realistic , protein expression profiles , such as that in Fig 4B . Even without such simplification , adopting empirical protein expression patterns for our simulation consistently supported the above results ( S10 Fig ) . We also note that the cuspidate waveforms in the plant clock do not simply result from the sampling intervals of experimental data , as different interpolation methods for these data points ( and the absence of such interpolation itself ) gave similar profiles . Provided that a cuspidate profile confers accurate timing of biological events around the peak phase , what is the implication of our simulation results involving the cuspidate waveform and inhibitory or activating regulation ? Inhibition-induced large phase differences between the genes correspond to the global coordination of multiple clock events , distant from each other in their peak times . Conversely , activation-induced small phase differences between the genes may coordinate only the clock events nearby in time . It is possible that activating regulation might also induce larger phase differences between the genes , but would not generate cuspidate profiles in this case ( S9 Fig ) . This fact explains why the kernel does not keep the activating regulations by REVEILLE 8 ( RVE8 ) , whose target genes have large phase differences with RVE8 , yet exhibit cuspidate profiles ( hence , those profiles are presumably more attributed to other regulators of these target genes ) . To summarize , inhibitory interactions in the plant clock seem to support the temporal coordination of distant clock events peaked at very specific times . However , it should be stressed that inhibitory interactions do not necessarily result in cuspidate waveforms in all cases . Rather , obtaining such waveform profiles requires inhibitory interactions when involving genes with large phase differences in their peak expression . Employing the terms in propositional logic , the presence of both cuspidate waveforms and large phase differences is close to a sufficient condition to implicate inhibitory regulation as their cause , but is not the necessary condition . We also note that our current definition of a cuspidate waveform is largely qualitative , based on a particular type of asymmetry between the acrophase and bathyphase . Mathematically more rigorous characterization , along with the inclusion of other possible waveforms in our framework , deserves investigation . Within the MF2015 kernel structure , a cuspidate-waveform gene which has multiple inhibitors tends to have larger phase differences with its strongest inhibitor , consistent with our framework . For example , the transcription of a cuspidate-waveform gene , PRR5 , is repressed by both LHY/CCA1 and TOC1 . There is a large phase difference between PRR5 and LHY/CCA1 proteins , ~8 h compared to the ~4 h difference between PRR5 and TOC1 proteins in a 12L:12D cycle . Supportively , MF2015 suggests that LHY/CCA1 inhibits PRR5 expression ~17 times more than TOC1 ( S1 Text ) . This fact indicates that the primary role of the PRR5 inhibition by LHY/CCA1 is to ensure the PRR5’s cuspidate waveform . Lowering the relative contribution of this inhibition ( i . e . , alleviating the repression by LHY/CCA1 while strengthening that by TOC1 ) reduces the peak-to-trough change in the PRR5 expression over time ( performed under 12L:12D cycles to control for the periods of different expression profiles; see S11 Fig ) . Our analysis accounts well for why PRR5 inhibition by LHY/CCA1 is present in the clock , although it is antagonistic to the system’s overall oscillatory capability as noted previously in relation to loop I . However , we recognize that it may be hard to treat separately multiple transcription factors regulating the same gene when considering their regulatory effects . Even in this case , we suggest that the combined activity profile of those transcription factors , which can be mapped into a mathematically equivalent single transcription factor’s profile , should follow our aforementioned condition when the target gene displays a cuspidate waveform . Generally , it is known that dynamical systems with activating interactions alone do not easily generate oscillations; inhibitory interactions are also necessary . Specifically , an odd number of inhibitions need to be arranged along a feedback loop , if the loop is not too long [40–42] . In addition to this basal level of inhibitory interactions required , an abundance of cuspidate-waveform genes in the plant oscillator tips the balance in favor of a greater number of inhibitory interactions , resulting in their dominance , according to our hypothesis [cuspidate protein profiles include LHY , PRR5 , TOC1 , EARLY FLOWERING 4 ( ELF4 ) , LUX ARRHYTHMO ( LUX ) , and GIGANTEA ( GI ) profiles in S1–S5 Figs , and comprise at least half of the available protein profiles . Among the corresponding genes , light-responsive genes are only LHY and GI ( Fig 1A ) , which yet maintain cuspidate expression patterns in LL and DD ( S4 and S5 Figs ) . It indicates that these patterns are largely independent of light stimulation] . For example , in loop I of the kernel , we note that both morning ( LHY/CCA1 ) and evening ( TOC1 ) genes show cuspidate profiles with a large phase difference between them , and are thus likely to require their own inhibitors . The simplest solution would be to have the two genes repressed by each other , but this solution , with an even number of inhibitions , would not generate oscillations . Hence , one more inhibitor , PRR5 , is necessary and the subsequent introduction of the double negative connection from TOC1 to LHY/CCA1 through PRR5 , combined with the TOC1 inhibition by LHY/CCA1 , completes the repressilator structure . In addition , PRR5 should maintain a large phase difference with LHY/CCA1 , because of the LHY/CCA1’s cuspidate profile . Consequently , PRR5 should show a small phase difference with TOC1 . Because of this small phase difference , the inhibition of PRR5 by TOC1 cannot alone produce the empirically-observed cuspidate PRR5 profile . Therefore , PRR5 requires an additional inhibitor with a large phase difference , LHY/CCA1 . The resulting inhibition of PRR5 by LHY/CCA1 now completes the full loop-I circuit . Through this analysis of loop I , the underlying mechanism of oscillatory dynamics with cuspidate waveforms was found to explain not only the prevalence of inhibitory interactions , but also the very specific , fine-resolution structure of loop I , revealing the loop’s organizing principle . Motivated by the intriguing connection between the shape of the waveforms and inhibitory regulation in plants , we asked if such relationships are observed in other circadian systems . Notably , a prevalence of inhibitory interactions per se is not conserved in other organisms: the core circadian systems of other organisms are usually simpler than those of plants , and involve feedback loops with comparable numbers of positive and negative interactions [11–13] . Those interactions are not necessarily transcriptional , and thus , caution should be taken when they are analyzed in our waveform-shape framework , which has been derived from the mathematical models of transcriptional regulation . Despite this caveat , in a preliminary analysis below , we applied our framework to both transcriptional and non-transcriptional interactions , considering their possible mathematical similarity at the coarse-grained level . In the core circadian clock of the fungus Neurospora crassa , WHITE COLLAR-1 , 2 ( WC-1 and WC-2 ) proteins form a WHITE COLLAR COMPLEX ( WCC ) that activates the expression of frequency ( frq ) gene . The expressed FRQ protein subsequently blocks the WCC activity by the clearance of WC-1 [12] . In this negative feedback loop , WC-1 is suppressed by FRQ , which is upregulated by WC-1 . From the experimental data [43] , we observed that WC-1 exhibits a cuspidate profile , when having a large phase difference ( ~11 hours in DD ) with FRQ . At the same time , FRQ shows a smooth sinusoidal profile . Despite multiple complicating factors in a rigorous analysis of species other than plants , this preliminary result from the Neurospora data is supportive of a relation between waveform-shape , phase differences , and interaction types ( activation or inhibition ) , which is suggested by our waveform-specifying framework .
In this study , we explored the underlying mechanism of the plant circadian system through a systematic in silico analysis of the clock gene circuitry , revealing its kernel architecture to be an interaction between four negative feedback loops dominated by inhibitory regulations ( Figs 1A and 2 ) . The kernel encompasses about half of the currently known interactions in the system , and they must be present to generate molecular rhythms close to WT . The other interactions not belonging to the kernel may play a role to improve the system’s robustness to diverse disturbances ( S1 Text ) , or may be required to form WT rhythms but under light conditions that have not been considered here due to limited data availability . A follow-up analysis is warranted for a more holistic understanding of plant circadian dynamics . Overall , our study illustrates the remarkable utility of mechanistic simulations , which can complement experimental approaches , in deciphering important biological processes [44–46] such as circadian rhythms . We suggested that a preponderance of inhibitory interactions at the core of the plant clock reflects abundant cuspidate profiles of clock genes , and facilitates the global coordination of temporally-distant clock events which are sharply peaked at very specific times . We envisage that this type of cuspidate waveforms helps confer high-resolution timing to many subsequent downstream tasks in plant physiology and development [35 , 47] . Whether a certain class of waveforms other than cuspidate shapes will also benefit from inhibitory interactions will be an interesting issue to address . Besides the effect on waveforms , alternative hypotheses might be possible to explain the prevalence of the inhibitory interactions , e . g . , in the context of stochasticity in molecular events , or the system’s response time [48–50] . Yet , we are not aware of any explicit link or evidence to connect those mechanisms to dominant inhibitory interactions in the plant clock . Nevertheless , the possible relevance of those mechanisms deserves active investigation , towards a comprehensive picture of the plant circadian system viewed from various angles . The four negative feedback loops within the kernel present an array of interesting predictions , which are experimentally testable . The Δprr5/prr7 double mutation severely impairs the free running rhythmicity of clock-controlled gene expression [36] . According to our prior discussion of the loops-I-to-III inactivation , only the removal of both PRR5 and PRR7’s inhibitory actions on LHY and CCA1 , rather than entire deletions of PRR5 and PRR7 , should suffice to phenocopy the double mutant , or at least , to considerably alter clock gene expression patterns . Additionally , from the reciprocal inhibitions within the unique loop-I structure , we suggested that an increase of the PRR5 inhibition by LHY/CCA1 would lengthen the free running period and that the opposite perturbation would shorten the period ( S1 Text ) . Furthermore , in the context of inhibitory interactions and cuspidate waveforms , we proposed that decreasing the PRR5 inhibition by LHY/CCA1 under 12L:12D cycles , balanced by strengthening the PRR5 inhibition by TOC1 , would reduce the peak-to-trough change in the PRR5 expression profile ( S1 Text ) . Experimental validation of all these predictions would require manipulation of specific interactions between genes , rather than the alteration or deletion of the functionality of the entire gene itself . This could be achieved , for example , by modifying key cis-regulatory elements at the relevant promoter sites . Any discrepancy between experimental and computational results might be useful for our model improvement . Further consideration of protein segregation into different cellular compartments [27 , 28] , stochastic fluctuation in mRNA and protein concentrations [49 , 51 , 52] , stimulus by temperature changes and endogenous sugar supply [53 , 54] , and tissue-specific clock regulation [55] offers additional avenues towards more complete mathematical models . Various methods to infer biological networks would also contribute to this direction [56–59] . Finally , our systematic approach advances the goal for a fundamental design principle of biological clockwork [53 , 60–62] , as well as for an optimal circuitry design in synthetic biology [32 , 63 , 64] .
We constructed our mathematical model ( MF2015 ) of the core circadian clock in Arabidopsis by applying system identification techniques [65] . Transcriptional , post-translational , and light regulations of molecular components were considered for model construction , primarily based on experimentally verified knowledge . The model consists of 24 ODEs employing Michaelis-Menten kinetics . Each ODE describes the concentration rate change of the corresponding mRNA , protein , or protein complex: typically , for mRNAs , ċm ( t ) = f1[{cTF ( t ) } , {h} , {θ}]–g1[cm ( t ) , {θ}] , and for proteins , ċp ( t ) = f2[cm ( t ) , {θ}]–g2[cp ( t ) , {θ}] . Here , cm ( cp ) denotes mRNA ( protein ) concentration , cTF denotes the transcription factor concentration , t is time , the function f1 ( f2 ) describes transcriptional ( translational ) mechanisms , the function g1 ( g2 ) describes mRNA ( protein ) degradation , θ’s are model parameters , h’s are the Hill coefficients , and {…} includes single or multiple elements . If experimental evidence indicates that transcription factors form a dimer , we set the Hill coefficient to be 2 , otherwise , it is set to 1 [33 , 66] . Transcriptional regulation in f1 is modeled by θ1 ( cTF ) h/[θ2h+ ( cTF ) h] for activation or θ1/[θ2h+ ( cTF ) h] for inhibition . The regulatory effect of multiple activators ( inhibitors ) is modeled by the summation ( product ) of individual regulatory effects , with some exceptions such as PRR proteins ( S1 Text ) [29 , 67] . We model the binding of ZTL and GI proteins by adapting the alternative Michaelis-Menten relation in [68] . For the model parameter estimation , we collected experimental time course data of mRNA and protein levels in WT Arabidopsis from publicly available sources listed in S1 Table . Because the absolute mRNA and protein levels were difficult to ascertain from their sources , we normalized the mRNA and protein levels into dimensionless values ( ≤1 ) with arbitrary scales ( S1 Text ) . We compared the simulation results with experimental data and applied the prediction error method with a quadratic criterion [65] to estimate the parameters; minimization of a mean squared error between the simulated and experimental data gave rise to the estimated parameters . Before the minimization , the initial parameters were chosen using a linear least square method described in [26] . The minimization was performed using the MATLAB function fminsearch . In cases where constraints need to be imposed on the parameters to avoid over-fitting or biologically unrealistic solutions , the MATLAB function fmincon was used . Full details of the model construction , equations , and parameters are presented in S1 Text . In this study , a kernel is defined as a collection of spanning subgraphs that satisfy the following condition: each spanning subgraph contains all molecular components in the system and a minimal subset of their regulatory interactions ( including light regulation ) , which are necessary to generate the temporal trajectory of molecular concentrations close to those of WT . Identification of the exact kernel demands very extensive computational resources; therefore , we used a heuristic approach to estimate the kernel structure . In this procedure , both molecular interactions and light regulations are referred to simply as interactions . First , we simulated the knockout effect of each interaction on WT expression patterns under five different light conditions . The knockout effect was quantified for each molecular component and light condition , by a root mean square error ( RMSE ) between the simulated mutant and WT expression profiles of the component in that light condition ( S1 Text ) . After deletion of a given interaction , we identified the largest value ( RMSEmax ) among RMSEs for all components and light conditions except for GI and ZTL proteins in LL ( RMSEGI , LL and RMSEZTL , LL ) . Based on our manual inspection , the model outputs appear to remain robust if they simultaneously satisfy RMSEmax ≤ 0 . 2 , RMSEGI , LL ≤ 0 . 5 , and RMSEZTL , LL ≤ 0 . 5 ( because GI and ZTL levels are substantially elevated in LL , they allow relatively large RMSEs ) . From MF2015 , we pruned all interactions with small knockout effects ( RMSEmax ≤ 0 . 2 , RMSEGI , LL ≤ 0 . 5 , and RMSEZTL , LL ≤ 0 . 5 ) . The simulated profiles with the only remaining interactions after the pruning still showed RMSEmax ≤ 0 . 2 , RMSEGI , LL ≤ 0 . 5 , and RMSEZTL , LL ≤ 0 . 5 . Among these remaining interactions , we focused on the interactions that satisfy RMSEmax ≤ 0 . 3 , RMSEGI , LL ≤ 0 . 8 , and RMSEZTL , LL ≤ 0 . 8 . We found that some of these interactions can be additionally removed from the system because the simultaneous deletion of those interactions eventually resulted in RMSEmax ≤ 0 . 2 , RMSEGI , LL ≤ 0 . 5 , and RMSEZTL , LL ≤ 0 . 5 , when parameter re-optimization was performed ( S1 Text ) . We did not attempt to delete interactions with larger RMSEmax , RMSEGI , LL , or RMSEZTL , LL ( RMSEmax > 0 . 3 , RMSEGI , LL > 0 . 8 , or RMSEZTL , LL > 0 . 8 ) with the original parameters , because these RMSEs were not usually reduced to RMSEmax ≤ 0 . 2 , RMSEGI , LL ≤ 0 . 5 , and RMSEZTL , LL ≤ 0 . 5 after parameter re-optimization . The exception to these procedures was the PRR7 inhibition by TOC1 . This interaction was removed initially because of small RMSEs caused by the deletion . In fact , the small RMSEs resulted from the PRR7 inhibition by the EC , which buffered the loss of the inhibition by TOC1 . The PRR7 inhibition by TOC1 and that by the EC are almost equivalent to each other , because of the same target gene ( PRR7 ) and regulation type ( inhibition ) , and similar TOC1 and EC profiles in MF2015 . Indeed , the removal of the PRR7 inhibition by the EC from MF2015 was compensated for by the inhibition by TOC1 when accompanied by parameter re-optimization . Because of the equivalence of these two inhibitory interactions , we reinstated the PRR7 inhibition by TOC1 in the kernel structure . No other interaction was reinstated due to a lack of such equivalence . The simulation of the resulting kernel structure with re-optimized parameters produces the WT expression profiles that capture the overall experimental and MF2015-simulated profiles ( S1–S5 Figs ) . Further details of the kernel identification are presented in S1 Text . To investigate how transcriptional regulation affects the formation of cuspidate profiles , we considered a mathematical system containing a single transcription factor ( either an inhibitor or activator ) and its own target gene ( Fig 4A ) . The ODEs for this system are given by x˙m ( t ) =g[xTF ( t ) , h , {α}]−λmxm ( t ) and x˙p ( t ) =xm ( t ) −λpxp ( t ) , where xTF denotes the transcription factor concentration , xm ( xp ) denotes the target gene’s mRNA ( protein ) concentration , t is time , g = xTFh/ ( α1+α2xTFh ) + α3 if the transcription factor is an activator , g = 1/ ( α1+α2xTFh ) if the transcription factor is an inhibitor , h is the Hill coefficient , and α’s and λ’s are constants . In the equation for x˙p ( t ) , without loss of generality , we omitted the coefficient for a protein synthesis rate per mRNA in front of xm ( t ) . Therefore , technically , xm ( t ) should be interpreted as the protein synthesis rate , rather than as the mRNA concentration itself . Although the equation for x˙m ( t ) was formulated for the case of a single transcription factor , it generally works for multiple transcription factors as well , because the combined activity profile of these transcription factors ( represented by g ) can be mapped into a mathematically equivalent single transcription factor’s profile . To generate xp ( t ) having a cuspidate waveform schematized in Fig 4B , we considered various forms of xTF ( t ) and activating and inhibitory regulations . Given the form of xTF ( t ) , we computed xp ( t ) with the parameters that best fit xp ( t ) into a cuspidate profile in Fig 4B . The resulting xp ( t ) was compared to Fig 4B , and their similarity was evaluated . Further details are presented in S1 Text . | Sleep/wake cycles in animals exemplify daily biological rhythms driven by internal molecular clocks , circadian clocks , which are important for plant life as well . The plant circadian clock is highly complex , eluding our understanding of its design principle . Based on the computational simulation of Arabidopsis thaliana , we successfully identified a kernel of the plant circadian system , the critical genetic circuitry for clock function . The kernel integrates four major negative feedback loops that process molecular circadian oscillations . Surprisingly , the plant clock circuitry was found to be overwhelmingly composed of inhibitory , rather than activating , interactions among genes . This fact underlies plant circadian molecular profiles to often exhibit sharply-shaped , cuspidate waveforms , which indicate clock events that are markedly peaked at very specific times of day . Our work presents experimentally-testable predictions , with implications for synthetic biology . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"gene",
"regulation",
"regulatory",
"proteins",
"messenger",
"rna",
"dna-binding",
"proteins",
"circadian",
"oscillators",
"transcription",
"factors",
"chronobiology",
"proteins",
"gene",
"expression",
"biochemistry",
"circadian",
"rhythms",
"genetic",
"oscillators",
"rna",
"biochemical",
"simulations",
"nucleic",
"acids",
"genetics",
"biology",
"and",
"life",
"sciences",
"computational",
"biology"
] | 2016 | Kernel Architecture of the Genetic Circuitry of the Arabidopsis Circadian System |
Among the arthropod-borne nematodes infesting dogs , Onchocerca lupi ( Spirurida: Onchocercidae ) is of increasing zoonotic concern , with new human cases of infection diagnosed in Turkey , Tunisia , Iran and the USA . Knowledge of the biology of this nematode is meagre . This study aimed at assessing the distribution and periodicity of O . lupi microfilariae from different body regions in naturally infested dogs . Skin samples were collected from six dogs infested with O . lupi but without apparent clinical signs . Two skin samples were collected from 18 anatomical regions of dog 1 at necropsy . In addition , single skin biopsies were performed from the forehead , inter-scapular and lumbar regions of dogs 2–6 , in the morning , afternoon , and at night . Two aliquots of the sediment of each sample were microscopically observed , microfilariae counted and morphologically and molecularly identified . Most of the 1 , 667 microfilariae retrieved from dog 1 were in the right ear ( 59 . 6% ) , nose ( 26 . 5% ) , left ear ( 6 . 7% ) , forehead ( 3 . 0% ) , and inter-scapular ( 2 . 9% ) regions . In dogs 2–6 , the overall mean number of microfilariae was larger on the head ( n = 122 . 8 ) , followed by the inter-scapular ( n = 119 . 0 ) and lumbar ( n = 12 . 8 ) regions . The overall mean number of microfilariae was larger in the afternoon ( 153 . 4 ) , followed by night ( 75 . 4 ) and morning ( 25 . 8 ) . Onchocerca lupi microfilariae were more common in the head ( i . e . , ears and nose ) than in the remaining part of the dog's body , indicating they tend to aggregate in specific body regions , which are the best sites to collect skin samples for diagnostic purposes . The periodicity pattern of microfilariae of O . lupi and their concentration in specific body regions is most likely a result of the co-evolution with their as-yet-unknown vector . The detection of skin microfilariae in asymptomatic animals , suggests the potential role of these animals as carriers and reservoirs of O . lupi .
Vector-borne nematodes of the family Onchocercidae ( Spirurida ) are of major medical concern . Among others , adult worms of Wuchereria bancrofti and Brugia malayi may live in the lymphatic system of humans causing obstruction ( i . e . , elephantiasis ) and those of Onchocerca volvulus in the subcutaneous tissues , with microfilariae inducing systemic or localized abnormal immune-mediated response , ultimately leading to severe ocular onchocercosis [1] . Some of these diseases may impact human health; for instance , the so-called “river blindness” caused by O . volvulus affects about 17 . 7 million people globally [2] . Among the arthropod-borne helminths of dogs , an increasing zoonotic role is recognized for Dirofilaria immitis and Dirofilaria repens , which are characterized by blood circulating microfilariae that may eventually infest the eyes and other organs of patients [3] . In contrast , data on the biology of onchocercid nematodes of the genera Onchocerca and Cercopithifilaria , characterized by subcutaneous localized microfilariae in dogs , is meagre [3] . Onchocerca lupi , a parasite of the connective tissue of sclera , has been sporadically reported in symptomatic dogs from Hungary , Greece , Germany and Portugal [4]–[7] and , more recently , also in dogs and cats from the USA [8] , [9] . In dogs , this filarial worm may cause ocular lesions ranging from no apparent clinical sings [10] to blindness [11] , with subconjunctival granulomas representing the finding most commonly reported [5] . A recent study on 107 dogs sampled in Greece and Portugal reveals an overall prevalence of infestation with O . lupi of 8 . 4% [10] . Since the first report of human ocular infestation by O . lupi [12] this parasite has been recognised as a zoonotic agent in patients from Turkey , Tunisia [13] , Iran [14] and the USA [15] . Despite the resurrected interest of scientific community towards this onchocercid , knowledge of its biology remains obscure and its vectors are still unknown . Therefore , this study aimed at assessing the distribution , abundance and periodicity of O . lupi microfilariae collected from different body regions in six naturally infested dogs .
The study was conducted according to the principles of Good Clinical Practice ( VICH GL9 GCP , 2000 http://www . ema . europa . eu/docs/en_GB/document_library/Scientific_guideline/2009/10/WC500004343 . pdf ) and procedures were approved by the Ethical commission at the University of Évora ( identification number: AE02Fila2013 ) as complying with the Portuguese legislation for the protection of animals ( Law no . 92/1995 , from 12th of September ) . An owner consent agreement was obtained before sampling collection . On March 2013 , skin samples were collected from six mongrel dogs ( i . e . , two males and four females ) , from four to 10 years of age , living in the municipality of Olhão , Algarve region , southern Portugal ( latitude 37°01′42″N , longitude 7°50′33″W , 8 meters above the sea level ) . All animals were previously identified as infested with O . lupi by the examination of skin snip sediments , during an epidemiological survey conducted in the study area [10]; none of the dogs had received endo- or ecto-parasitic treatments . One of the dogs ( dog 1 ) accidentally died due to gastric volvulus and , during necropsy , two skin samples were collected from 18 anatomical regions ( about 2 cm apart ) , distributed throughout the body surface ( Table 1 ) . In addition , single skin biopsies were performed from the remaining five dogs ( i . e . , dog 2–6 ) , from three anatomical regions ( i . e . , forehead , inter-scapular and lumbar regions ) at different time points ( i . e . , in the morning at 10:00 , late afternoon at 18:00 , and during the night at 23:00 h ) . All skin samples from the six dogs were collected using biopsy punches ( 4 mm in diameter ) and soaked in 2 ml saline solution ( NaCl 0 . 9% ) before observation . For each sample , two aliquots ( 20 µl each ) of the sediment were used to prepare temporary mounts , covered by an 18×18 mm coverslip , which were observed under a light microscope . Microfilariae were identified according to their morphology [11] , [16] . Briefly , O . lupi microfilariae are characterised by an unsheathed body 110 . 1 µm ± 7 . 5 SD long and 6 . 8 µm ± 1 . 2 SD wide , rounded anterior extremity bearing a tiny tooth and a bent tail 11 . 7 µm long Figure 1 . Additionally , three biopsy punches ( 8 mm in diameter ) were taken from the nose and the peri-ocular regions at the necropsy of dog 1 for histological examination ( see below ) . These skin samples were fixed in 4% buffered formalin solution ( pH 7 . 4 ) , embedded in paraffin and routinely processed for light microscopy . Thick sections ( 5 µm ) were stained with haematoxylin and eosin before being microscopically observed . The morphological identification was confirmed by molecular amplification and sequencing of the partial cytochrome oxidase subunit 1 ( cox1 ) gene , following procedures described elsewhere [13] . Nucleotide sequences , examined by BLAST tool , displayed 100% homology with sequences of O . lupi from Portugal deposited in GenBank ( accession number: EF521410 ) . Skin samples from dog 1 were soaked in saline solution for approximately 6 h ( first replicate ) and 12 h ( second replicate ) before observation , whereas samples from dog 2–6 were counted in a single assessment within 12 h after collection . The mean number ( ± standard deviation ) of microfilariae was calculated according to body location and periodicity . Data normality was assessed using Lilliefors test and then the mean number of microfilariae according to collection site and period was compared using one way ANOVA , with Tukey post hoc test or Mann-Whitney U test as appropriate . A p<0 . 05 was considered statistically significant . Statistical analysis was conducted using BioEstat ( version 5 . 0; Mamiraua/CNPq , Belem , PA , Brazil ) .
All sampled animals were apparently healthy , presenting no apparent ocular alteration . The number of O . lupi microfilariae from each body site assessed at the necropsy of dog 1 is reported in Table 1 . A total of 1 , 667 microfilariae of O . lupi were collected , most ( 95 . 8% ) of which from the head . In particular , most of the microfilariae were located in the right ear ( 59 . 6% ) , nose ( 26 . 5% ) , left ear ( 6 . 7% ) , forehead ( 3 . 0% ) , and inter-scapular ( 2 . 9% ) regions . Only 21 microfilariae ( 1 . 3% ) were found in the remaining regions of the dog's body . Of the 12 body regions that resulted positive for microfilariae , eight were positive at both replicates ( Table 1 ) , with a higher percentage of skin samples positive at the examination of the first aliquot ( n = 20; 71 . 4% ) than of the second ( n = 8; 28 . 6%; data not shown ) . Accordingly , the mean number of microfilariae counted in the first aliquot was higher than in the second ( Mann-Whitney U test , p = 0 . 02 ) , with up to 825 O . lupi microfilariae counted in a single sample from the right ear of dog 1 ( Figure 2 ) . The overall number of microfilariae retrieved in samples after 12 h of soaking ( second replicate ) was over 5 times higher than that after 6 h of soaking ( first replicate ) . However , no significant difference was found in relation to the mean number of microfilariae/µl counted in each body site in the first and second replicates ( Mann-Whitney U test , p = 0 . 37 ) . Microfilariae were alive at both assessments . In dogs 2–6 , the mean number of microfilariae was higher on the head ( 40 . 9±35 . 0 ) , followed by inter-scapular ( 39 . 7±34 . 6 ) and lumbar ( 4 . 3±2 . 7 ) regions ( Figure 3a ) ; however , no statistically significant difference was found in relation to body site ( ANOVA , p = 0 . 11 ) . The mean number of microfilariae per body site varied among dogs 2–6 , with some dogs presenting more microfilariae in the head and others in the inter-scapular region ( Figure 3b ) . As far as periodicity , the mean number of microfilariae was larger in the late afternoon ( 51 . 1±28 . 5 ) , followed by night ( 25 . 1±4 . 7 ) and morning ( 8 . 6±8 . 0 ) ( Figure 4a ) . Indeed , the mean number of microfilariae found in the morning sampling was significantly lower than that found in the late afternoon ( ANOVA , p<0 . 01; Tukey post hoc test , p<0 . 01 ) . Interestingly , the peak of microfilariae occurred during the night in dog 2 ( Figure 4b ) . A few slender microfilariae were detected on histopathological examination of the peri-ocular regions in the dermis . They were unevenly distributed into the connective tissue among fibres in the perifollicular and interfollicular areas and in the deep dermis in the proximity of small vessels ( Figure 5 ) . Skin samples showed dermatitis with mild superficial and periadnexal perivascular infiltrates composed of eosinophils and a few lymphocytes . Inflammatory changes were accompanied by hyperplasia and ortokeratotic hyperkeratosis with a few coccoid bacteria between corneocytes .
Until now , information on the distribution and abundance of O . lupi microfilariae in the skin of infected dogs was limited to a single report on four symptomatic dogs sampled at the periocular and umbilical areas [17] . Interestingly , in spite of the absence of apparent clinical signs , animals were positive for skin microfilariae , suggesting the potential role of these animals as asymptomatic carriers and reservoirs of O . lupi . Data on the distribution of O . lupi microfilariae ( dog 1 ) showed they are more abundant in the head ( i . e . , ears and nose ) than in the remaining part of the dog's body . Although this pattern was confirmed by the data on larval periodicity ( dogs 2–6 ) , in the latter case , the difference between the mean number of microfilariae from forehead and inter-scapular regions was not significant . This might be due to the fact that in dogs 2–6 , skin was sampled from the forehead area , which , in turn , displayed a smaller number of microfilariae in comparison with ears and nose ( Table 1 ) . Consequently , the diagnosis of O . lupi infestation in dogs should be performed via the examination of skin samples collected from the ears or nose . Nonetheless , the inter-scapular region might be a preferred site because some dogs ( dogs 2 , 4 and 5 ) actually presented more microfilariae in the inter-scapular region in comparison with the forehead . Furthermore , the inter-scapular region is less vascularized and better accepted by both animals and owners as a site to be biopsied . In addition , this site might be more practical to be sampled during large population surveys [10] . The results of this study contrast with previous data on the distribution of O . lupi microfilariae [17] , in which only peri-ocular and umbilical regions were considered as preferential sites for skin snipping . Although some authors [17] quantified the larval concentration ( i . e . , 267 . 5 larvae per gram of skin ) , they did not report the exact amount of skin tissue sampled and , most likely , underestimated the number of microfilariae due to the short period of soaking they adopted ( i . e . , 1 h ) . Indeed , the increased number of microfilariae at the second assessment ( 12 h ) suggests that the longer is the duration of soaking , the highest is the probability to find microfilariae in the skin sediment . From a diagnostic perspective , microfilariae of O . lupi should also be differentiated from those of other filarial nematodes ( i . e . , Cercopithifilaria bainae , Cercopithifilaria grassii and Cercopithifilaria sp . II sensu Otranto et al . ( 2011 ) , which may be retrieved at the same time in the dermis of dogs [18] . As recorded for other species within the genus , it becomes evident that O . lupi microfilariae tend to aggregate in specific body regions ( i . e . , head and inter-scapular region ) . This might be determined by the proximity of gravid O . lupi females to the sampling sites , as indicated in previous reports [11] . Similarly , microfilariae of Onchocerca gutturosa , Onchocerca ochengi , and Ochocerca dukei of cattle are found on dorsal side , posterior declivous abdomen and navel , respectively , in the same regions where adults are found [19] . Nonetheless , the adult localization of Onchocerca spp . is not mandatorily related to the preferred area of microfilariae localization , because the latter might migrate far away from females , through the lymphatic system [20] , [21] . This is the case of microfilariae of Onchocerca tarsicola parasitizing red deer ( Cervus elaphus ) , which are concentrated mainly in the external ears , whereas adults are present in the radial-carpal and tibia-tarsal joint tendons [22] . The histological evidence of mild skin eosinophilic inflammation nearby the microfilariae might be also a non-specific finding , not necessarily associated with the presence of the O . lupi larvae . Indeed , eosinophilic inflammation is usually seen in allergic as well as in parasitic skin diseases , thus other causes for such a condition cannot be ruled out . In addition , the minimal inflammatory response to O . lupi microfilariae could also be due to the fact they were recently released from a gravid female , as suggested for microfilariae of D . immitis [23] . The concentration of microfilariae of O . volvulus on the hip , shoulders and lower parts of the body [24] coincide with the sites where its black fly vectors ( e . g . , Simulium damnosum sensu lato ) preferentially feed [25] . Undoubtedly , the occurrence of O . lupi microfilariae in specific body regions is most likely a result of the co-evolution between competent vectors , hosts and the parasite . Indeed , circadian variations of microfilariae reported in filarial worms with blood circulating microfilariae ( e . g . , D . immitis , D . repens , Loa loa , and Wuchereria bancrofti ) is considered to be an adaptation to the biting behaviour of the vectors , the circadian rhythms of the host and to variations in environmental temperature and humidity [26] . This pattern has also been demonstrated for filarial worms with subcutaneous microfilariae as those of O . volvulus , in which the maximal larval density overlaps the peak of activity ( i . e . , between 18:00 and 19:00 ) of its Simulium vector [27] , [28] . In the case of O . lupi , in absence of any scientific evidence , the role of mosquitoes ( e . g . , Culex pipiens pipiens , Anopheles spp . ) , or of biting midges species as vectors cannot be ruled out [3] . However , blackflies , whose biting activity increases in late morning or early afternoon [25] , remain a major candidate as a vector of O . lupi . For example , Simulium reptans , a species collected where O . lupi cases have been reported in dogs ( i . e . , Germany , Greece , Hungary , Portugal and Switzerland ) [29] displays exophilic and exophagic behaviours , with the highest biting activity during the afternoon [30] . Although suspected , the vector role of S . reptans has never been ascertained [29] . The information on the distribution and periodicity pattern of microfilariae of O . lupi here reported is of relevance not only for the comprehension of its biology , but also for a more refined diagnosis of the infestation . Indeed , in absence of any other diagnostic tool the “skin snip” remains the only option to detect larval stages in the subcutaneous tissue of infested dogs . Therefore , veterinary practitioners should be aware about the best body sites and period of the day for performing skin biopsy , in order to achieve a more reliable diagnosis toward a better comprehension of this little known parasite of increasing veterinary and medical concern . | Onchocerca lupi is a little known arthropod-borne helminth infesting dogs of increasing interest to the scientific community due to the recent demonstration of its zoonotic potential . Nonetheless , knowledge of the biology of this nematode is exiguous . In this study the distribution and periodicity of O . lupi microfilariae was investigated from different body regions in naturally infested dogs . Data indicate that O . lupi microfilariae were more common in the head ( i . e . , ears and nose ) followed by the inter-scapular region than in the remaining part of the dog's body suggesting that these parasites aggregate in these anatomical sites . These regions might be the best sites to collect skin samples for diagnostic purposes . Finally , the periodicity pattern of microfilariae of O . lupi and their concentration in specific body regions is most likely a result of the co-evolution with their as-yet-unknown vector . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] | [] | 2013 | Cutaneous Distribution and Circadian Rhythm of Onchocerca lupi Microfilariae in Dogs |
Fatal familial insomnia ( FFI ) and a genetic form of Creutzfeldt-Jakob disease ( CJD178 ) are clinically different prion disorders linked to the D178N prion protein ( PrP ) mutation . The disease phenotype is determined by the 129 M/V polymorphism on the mutant allele , which is thought to influence D178N PrP misfolding , leading to the formation of distinctive prion strains with specific neurotoxic properties . However , the mechanism by which misfolded variants of mutant PrP cause different diseases is not known . We generated transgenic ( Tg ) mice expressing the mouse PrP homolog of the FFI mutation . These mice synthesize a misfolded form of mutant PrP in their brains and develop a neurological illness with severe sleep disruption , highly reminiscent of FFI and different from that of analogously generated Tg ( CJD ) mice modeling CJD178 . No prion infectivity was detectable in Tg ( FFI ) and Tg ( CJD ) brains by bioassay or protein misfolding cyclic amplification , indicating that mutant PrP has disease-encoding properties that do not depend on its ability to propagate its misfolded conformation . Tg ( FFI ) and Tg ( CJD ) neurons have different patterns of intracellular PrP accumulation associated with distinct morphological abnormalities of the endoplasmic reticulum and Golgi , suggesting that mutation-specific alterations of secretory transport may contribute to the disease phenotype .
Prion strains with unique self-templating and neurotoxic properties are thought to emerge spontaneously in humans carrying genetic prion disease-associated PrP mutations , dictating the phenotypic expression of disease . Here we report that transgenic ( Tg ) mice carrying the PrP mutation associated with one of these diseases ( fatal familial insomnia , FFI ) develop severe sleep disorders and other key phenotypic features of the human disease , different from those seen in analogously generated Tg mice expressing another prion disease-associated mutation ( Creutzfeldt-Jakob disease , CJD ) . No prion infectivity is spontaneously generated in these mice , indicating that mutant PrP has disease-encoding properties that do not depend on self-templating competence . Prion diseases are progressive and invariably fatal degenerative disorders of the central nervous system ( CNS ) that affect humans and other animals [1] . CJD , FFI and Gerstmann-Sträussler-Scheinker ( GSS ) syndrome are the most common forms in humans; scrapie of the goat and sheep , bovine spongiform encephalopathy , and chronic wasting disease of deer and elk are the best-known prion zoonoses [2] . Neuronal loss , gliosis , spongiform change ( vacuolation of the neuropil in the gray matter ) and in some cases amyloid deposits are typical neuropathological findings in prion diseases , which in humans usually present with loss of motor coordination and other motor abnormalities , dementia and neurophysiological deficits . Similarly to other progressive neurodegenerative disorders , such as Alzheimer’s disease ( AD ) and Parkinson’s disease ( PD ) , frontotemporal dementia and the tauopathies , prion diseases can arise sporadically or be genetically inherited; however , they can also be acquired by infection [3 , 4] . The infectious agent ( prion ) is scrapie prion protein ( PrPSc ) [5] . This is a conformationally altered and aggregated isoform of the cellular prion protein ( PrPC ) which propagates by imprinting its aberrant conformation onto native PrPC molecules [6] . Genetic prion diseases are linked to point mutations and insertions in the PRNP gene encoding PrPC on chromosome 20 [7] . Point mutations are mainly clustered in the protein’s C terminus , leading to amino acid substitutions or protein truncations . The insertions consist of additional copies of an octapeptide repeat in the N-terminal region , which normally contains one nonapeptide and four octapeptides . Mutant PrP is thought to misfold and aggregate spontaneously , eventually acquiring the PrPSc structure . Different PRNP mutations are associated with distinct clinical and neuropathological phenotypes: CJD , FFI , GSS , PrP-cerebral amyloid angiopathy [7] and a recently described PrP systemic amyloidosis [8] . The disease phenotype is also influenced by PRNP polymorphic codon 129 , where either methionine ( M ) or valine ( V ) can be encoded . A typical example is the prion disease linked to the substitution of aspartic acid ( D ) to asparagine ( N ) at codon 178 , which , depending on the aminoacid encoded at polymorphic site 129 , segregates with either FFI ( D178N/M129 ) , primarily characterized by severe sleep disorders and autonomic dysfunction , or CJD178 ( D178N/V129 ) , clinically identified by global cortical dementia and motor abnormalities [9] . The reason for this variability is not known . There is evidence that D178N/M129 and D178N/V129 PrPs differ in their folding and supramolecular assembly [10–12] , but how conformational variants of the PrP polypeptide produce different diseases is not clear . Only recently have we begun to understand how mutant PrP causes neurological dysfunction . PG14 , a mouse ( mo ) PrP carrying a nine-octapeptide repeat insertion associated with GSS [13] , and moPrP D177N/V128 , homologous to the human CJD178 mutation , are partially retained in the neuronal endoplasmic reticulum ( ER ) [14 , 15] . Intracellular accumulation of these mutants impairs the secretory transport of the voltage-gated calcium channel ( VGCC ) α2δ-1 subunit , resulting in inefficient targeting of the VGCC complex to presynaptic terminals . This leads to inefficient glutamatergic neurotransmission in cerebellar granule neurons ( CGNs ) and abnormal motor behavior in Tg mice [15] . Thus in mouse models of GSS and CJD178 , ER retention of mutant PrP causes motor disease by altering the secretory trafficking of calcium channels essential for synaptic activity . To further explore the mechanisms of mutant PrP neurotoxicity and , specifically , the role of the 129 polymorphism in directing the disease phenotype , we developed a mouse model of FFI . Here we describe Tg ( FFI ) mice expressing moPrP D177N/M128 , which presented abnormalities in sleep-wake patterns and other pathological features highly reminiscent of FFI . Neurons in Tg ( FFI ) mice accumulate mutant PrP in the Golgi and show morphological alterations of this transport organelle . This suggests that different mutant PrPs may have different effects on secretory transport , potentially inducing specific functional abnormalities in neurons , hence clinically defined neurological diseases .
We produced Tg mice expressing moPrP D177N/M128 with or without an epitope tag for monoclonal antibody 3F4 . We identified ten founders ( four with and six without the 3F4 epitope ) . To generate the transgenic lines , referred to as Tg ( FFI ) , founders were bred with PrP knockout mice ( Prnp0/0 ) , so that the progeny expressed only mutant PrP . We established five Tg lines: one expressing 3F4-tagged ( FFI-K5 ) and four untagged mutant PrP ( FFI-10 , FFI-15 , FFI-26 and FFI-31 ) . Transgene copy number and mutant PrP expression are shown in Table 1 and Fig 1A and 1B . Western blot analysis showed that unglycosylated PrP was under-represented ( Fig 1 ) , like in humans carrying the D178N mutation [16] . Mutant PrP in the mouse brain was largely insoluble ( seen in pellet fractions in Fig 1C and 1D ) , and weakly protease-resistant ( Fig 1E , lanes 5–8 ) . After deglycosylation with PNGaseF , the PK-resistant fragment had an apparent molecular mass of 19 kDa ( Fig 1F , lane 2 ) , consistent with observations in FFI patients [16] . The PK-resistant fragment of D177N/V128 PrP in Tg ( CJD ) mice was also 19 kDa ( Fig 1F , lane 4 ) , different from mutant PrP from CJD178 patients , which has a PK-resistant core of 21 kDa [16] , confirming our previous observations [14] . Detergent-insoluble and PK-resistant PrP was already detectable in 50 days old mice ( S1 Fig ) , well before they developed clinical disease ( see below ) . As disruption of sleep is a key feature of FFI , we analyzed the sleep-wake patterns in Tg ( FFI ) mice . We used Tg ( FFI-26 ) /Prnp0/0 mice , which express the mutant protein at approximately twice the wild-type ( WT ) level ( Table 1 and Fig 1B ) and develop a fatal neurological syndrome with motor and cognitive deficits ( see below ) . To assess the effect of co-expression of WT PrP , we also analyzed Tg ( FFI-26 ) /Prnp+/0 mice , in which one Prnp allele was reintroduced by backcrossing Tg ( FFI-26 ) /Prnp0/0 with C57BL/6J mice ( hereafter referred to as non-Tg/Prnp+/+ ) . Circadian organization of sleep and motor activity was lost in Tg ( FFI ) /Prnp0/0 mice . As expected in nocturnal animals , non-Tg/Prnp+/+ mice slept ( summing up NREM and REM sleep ) about twice as long during the day than during the night ( 2 . 1 ± 0 . 2 times ) . This was also true for non-Tg/Prnp0/0 , and Tg ( FFI ) /Prnp+/0 mice , which slept 2 . 0 ± 0 . 1 and 2 . 4 ± 0 . 3 times more during the day than during the night , respectively . In contrast , Tg ( FFI ) /Prnp0/0 mice slept only fifty percent more ( 1 . 5 ± 0 . 1 times; p < 0 . 05 by one-way ANOVA , F3 , 31 = 5 . 672 ) . The disorganization of sleep circadian rhythms in Tg ( FFI ) /Prnp0/0 mice was confirmed by analysis of gross body movements . Non-Tg/Prnp+/+ , non-Tg/Prnp0/0 and Tg ( FFI ) /Prnp+/0 mice moved almost three to almost four times more during the night than during the day , respectively ( 2 . 9 ± 0 . 3 , 3 . 9 ± 0 . 4 and 3 . 7 ± 0 . 6 times ) . Tg ( FFI ) /Prnp0/0 mice moved only 1 . 6 ± 0 . 2 times more during the night than during the day ( p < 0 . 001 by one-way ANOVA , F3 , 31 = 10 . 819 ) . Sleep continuity and organization were affected in Tg ( FFI ) /Prnp0/0 mice . The number of transitions between different behavioral states ( an indicator of broken sleep ) was greater in Tg ( FFI ) /Prnp0/0 mice than in non-Tg/Prnp+/+ and non-Tg/Prnp0/0 mice , during both the light and dark phases ( Fig 2 ) . In Tg ( FFI ) /Prnp+/0 mice there were more transitions only in comparison to non-Tg/Prnp+/+ mice , and only during the light phase ( Fig 2 ) . In 8 out of 9 Tg ( FFI ) /Prnp0/0 mice , entry into REM sleep was abnormal . Tg ( FFI ) /Prnp0/0 mice entered REM sleep directly from wakefulness in 24 . 6 ± 6 . 5% of REM epochs , instead of going through NREM sleep , as normally occurs ( Fig 3 ) . This was never observed in the other groups of mice . Slow-wave activity ( SWA ) during NREM sleep ( a measure of sleep drive and depth [17] ) and EEG spindles ( which characterize this sleep phase ) were reduced in Tg ( FFI ) /Prnp0/0 mice . Although the amount of NREM sleep was not reduced in Tg ( FFI ) /Prnp0/0 mice in comparison to the other mice , NREM sleep SWA was significantly less in Tg ( FFI ) /Prnp0/0 mice than in both non-Tg/Prnp+/+ and non-Tg/Prnp0/0 mice during the first portion of the light phase ( Fig 4 ) . The density of EEG spindles during NREM sleep in the light phase was lower in Tg ( FFI ) /Prnp0/0 mice ( 38 . 4 ± 7 . 5 spindles/h ) than in both non-Tg/Prnp+/+ and non-Tg/Prnp0/0 mice ( 184 . 4 ± 11 . 0 and 148 . 0 ± 9 . 1 spindles/h , respectively; p < 0 . 01 by one-way ANOVA F3 . 31 = 35 . 587 ) . The density of spindles in Tg ( FFI ) /Prnp+/0 mice ( 106 . 8 ± 14 . 3 spindles/h ) was intermediate between that of non-Tg/Prnp+/+ and Tg ( FFI ) /Prnp0/0 mice , and significantly different from both . Besides starting abnormally , in Tg ( FFI ) /Prnp0/0 mice REM sleep differed in amount ( Fig 4 ) , and its EEG power in the theta ( 6–9 Hz ) band ( the EEG hallmark of rodent REM sleep ) was significantly lower than in the other groups . The loss of REM sleep in Tg ( FFI ) /Prnp0/0 mice was due to a reduction in the number of REM sleep bouts , but not their duration ( Fig 2 ) . During the light phase , REM sleep theta power was 11 . 8 ± 0 . 4% ( of REM sleep EEG total power ) in Tg ( FFI ) /Prnp0/0 mice , 13 . 7 ± 0 . 4% in non-Tg/Prnp+/+ , and 13 . 2 ± 0 . 3% in non-Tg/Prnp0/0 mice ( mean ± SEM; p < 0 . 05 and p < 0 . 001 , Tg ( FFI ) /Prnp0/0 vs . non-Tg/Prnp+/+ and non-Tg/Prnp0/0 , respectively; one-way ANOVA with Bonferroni’s correction ) . During the dark phase , REM sleep theta power was significantly lower in both Tg ( FFI ) /Prnp0/0 and Tg ( FFI ) /Prnp+/0 than in non-Tg/Prnp+/+ mice ( 11 . 8 ± 0 . 5% , 11 . 9 ± 0 . 5% and 13 . 9 ± 0 . 5% respectively; p < 0 . 05 ) . Since additional alterations may become apparent when the sleep drive is increased , we investigated the response to sleep deprivation . Mice were kept awake during the first 6 h of the light phase by gentle handling , then allowed to sleep freely in the following 18 h . REM and NREM sleep , shown in Fig 5 , was calculated hourly for each animal as the difference between the time spent in REM or NREM sleep during and after sleep deprivation , and the amount spent in the corresponding hour during baseline conditions ( undisturbed ) . The hour-by-hour differences were then summed to give a cumulative curve . Nontransgenic and Tg ( FFI ) /Prnp+/0 mice lost the same amount of REM sleep during deprivation ( Fig 5A ) . By the end of the recording period ( i . e . 18 h after the end of the sleep deprivation period ) , non-Tg/Prnp+/+ and non-Tg/Prnp0/0 mice fully recovered the REM sleep lost , whereas Tg ( FFI ) /Prnp+/0 mice did not ( Figs 4 and 5A ) . During sleep deprivation , Tg ( FFI ) /Prnp0/0 mice lost less REM sleep than all other mouse lines , because REM sleep was already markedly reduced in these mice in basal conditions ( Fig 4 ) . In the next 18 h , Tg ( FFI ) /Prnp0/0 mice slept as much as in undisturbed conditions , having little loss of REM sleep to recover ( Figs 4 and 5A ) . Mice of all genotypes lost the same amount of NREM sleep during sleep deprivation . During the next 18 h , they did not fully recover ( Fig 5B ) and compensated the loss of NREM sleep with an increase in the power of the EEG delta band in the first 6 h of recovery ( Fig 4 ) , as previously shown for rodents deprived of sleep by gentle handling for a short time [18] . EEG activity was altered in Tg ( FFI ) /Prnp0/0 and Tg ( FFI ) /Prnp+/0 mice . This consisted of bursts of high-voltage polyphasic complexes , similar to those described in Tg ( CJD ) mice expressing D177N/V128 PrP [14] , with frequency peaking at about 7 Hz . This activity was almost equally distributed during the light and dark parts of the light/dark cycle . In Tg ( FFI ) /Prnp0/0 mice polyphasic complexes were present in respectively 8 . 8 ± 3 . 4% and 12 . 0 ± 3 . 7% of the epochs of the light and dark parts of the light-dark cycle . In Tg ( FFI ) /Prnp+/0 polyphasic complexes were present in 0 . 60 ± 0 . 24% and 1 . 15 ± 0 . 60% of epochs of the light and dark phases . Polyphasic complexes were present in 9 . 6 ± 3 . 1% , 2 . 1 ± 1 . 0% and 16 . 4 ± 5 . 2% of epochs scored respectively as REM sleep , NREM sleep and wakefulness , in Tg ( FFI ) /Prnp0/0 mice . These percentages were 0 . 4 ± 0 . 2% , 0 . 5 ± 0 . 3% and 1 . 1 ± 0 . 5% in Tg ( FFI ) /Prnp+/0 mice . No pathological activity was detected in non-Tg mice . Tg ( FFI ) mice had progressive neurological disease . They developed ataxia , with abnormal flexed posture of the hind legs , kyphosis , and foot clasp reflex ( S2 Fig ) . The phenotype was evident from 202 ± 2 days ( mean ± SEM , n = 58 ) in Tg ( FFI-26 ) mice . As the disease progressed the mice lost weight , and were killed when unable to feed themselves , at 436 ± 4 days ( n = 87 ) . Prospective observation of individual mice indicated an average duration of the illness of 262 ± 4 days ( n = 32 ) . To check the earliest appearance of motor dysfunction , Tg ( FFI-26 ) mice were tested on the accelerating Rotarod . They performed well until 90 days of age , indicating normal development of motor function . From 110 days on , however , the mutant mice showed a significantly shorter latency to fall than nontransgenic littermates; their performance worsened with aging until they became unable to stay on the rod ( Fig 6A ) . A similar but less aggressive neurological illness was seen in Tg ( FFI-10 ) mice expressing mutant PrP at lower levels . Neurological signs were evaluated at a single time in a cohort of hemi- and homozygous Tg ( FFI-10 ) littermates of different ages . There was no disease in hemizygous Tg ( FFI-10 ) mice younger than 328 days , but 23 out of 37 ( 62% ) mice between 328 and 757 days showed mild neurological disease . Prospective observation indicated that these mice never reached a debilitating stage and survived as long as non-Tg littermates . In addition , some hemizygous Tg ( FFI-10 ) animals remained free of neurological signs , suggesting incomplete penetrance when the mutant PrP was expressed at wild-type levels . In contrast , all homozygous Tg ( FFI-10 ) mice older than 290 days had neurological disease , and reached a terminal stage at 698 ± 26 days ( n = 14 ) , indicating a profound effect of transgene zygosity on the manifestation and time course of the illness . Confirming this , the Rotarod task showed age-dependent , transgene-dose-related motor dysfunction in Tg ( FFI-10 ) mice ( Fig 6B ) . The non-breeding Tg ( FFI-28 ) founder expressing PrP at ~5X died with neurological symptoms at 477 days , whereas Tg ( FFI ) lines with mutant PrP levels below 1X ( Table 1 ) never developed neurological disease . Thus the appearance of neurological illness and its rate of progression were correlated with the expression level of mutant PrP , similar to other mouse models of genetic prion disease [13 , 14 , 19–21] . Co-expression of WT PrP had no effect on motor dysfunction , in contrast to the mitigating effect on the sleep abnormalities . For example , in a group of littermates consisting of 9 Tg ( FFI-26 ) /Prnp0/0 , 10 Tg ( FFI-26 ) /Prnp+/0 and 8 Tg ( FFI-26 ) /Prnp+/+ mice , symptom onset was at 207 ± 13 , 216 ± 12 and 220 ± 11 days respectively ( F2 , 24 = 0 . 29; p = 0 . 75 by one-way ANOVA ) . There were also no differences in Rotarod performance ( Fig 6C ) . We found alterations in long-term recognition and spatial working memory too in Tg ( FFI ) mice , tested in the novel object recognition task and eight-arm radial maze . To avoid confounding effects due to the motor deficit that develops in older mice , we tested Tg ( FFI-26 ) animals younger than 100 days . Mice were impaired in long-term memory , as shown by the lower discrimination index in the object recognition task compared to non-Tg mice ( Fig 7A ) . They also performed poorly in the eight-arm radial maze , which tests spatial working memory , making significantly more errors in the first eight training trials than controls ( Fig 7B ) . Latency to complete the test was longer in Tg ( FFI ) than non-Tg mice ( Fig 7C ) . This may reflect an impairment in choice-making , since there were no significant differences in the number of total movements in the open field ( total line crossings: non-Tg = 304 ± 13; Tg ( FFI-26 ) = 283 ± 23; mean ± SEM ) , confirming no motor deficit at this stage . We used magnetic resonance imaging ( MRI ) to investigate the effects of the FFI mutation on brain structure . There were no significant differences between the whole-brain volumes of Tg ( FFI-26 ) and non-Tg littermates at 80 days of age ( non-Tg = 464 ± 6 mm3; Tg ( FFI-26 ) = 451 ± 5 mm3; mean ± SEM , n = 5–6; p = 0 . 0823 by Mann-Whitney test ) . In Tg ( FFI-26 ) mice older than 400 days , the whole-brain volume was 12% smaller than controls ( non-Tg = 493 ± 4 mm3; Tg ( FFI-26 ) = 436 ± 6 mm3; mean ± SEM , n = 9; p < 0 . 0001 by Mann-Whitney test ) . Analysis of individual brain areas showed that the thalamic and cerebellar volumes were significantly smaller in Tg ( FFI-26 ) mice than in non-Tg littermates , but there were no differences in the other brain regions ( Fig 8 ) . Neuropathological examination of Tg ( FFI ) mice showed PrP deposition in the form of diffuse “synaptic-type” immunoreactivity in several regions . These deposits were most prominent in the entorhinal and pyriform cortex , cingulate gyrus , hippocampal formation , thalamus , caudatum , putamen , amygdala and the molecular layer of the cerebellar cortex ( Fig 9 ) . Synaptic-type deposits were also present in the other cortical areas and the septum . Dot-like and small round PrP-immunoreactive profiles were seen in several subcortical structures , including the stria terminalis , fimbria and thalamus ( Fig 9 ) . Strongly immunoreactive fiber tracts were observed in the stria terminalis and in the stratum lucidum of CA3 , corresponding to the hippocampal mossy fibers . Diffuse intraneuronal PrP immunoreactivity was present in the mesencephalic trigeminal nucleus , the vestibular nucleus and the lateral dorsal nucleus of the thalamus . The neurons of the mesencephalic trigeminal nucleus were outlined by dot-like immunoreactive profiles that were also scattered in the neuropil ( Fig 9J ) . The PrP deposits were not fluorescent after thioflavin S staining , indicating that they did not contain amyloid fibrils . No spongiform-like changes were seen . Immunohistochemistry with the anti-GFAP antibody revealed astrogliosis mainly in the hippocampus , external layer of the cerebral cortex , and cerebellum ( S3A-S3F Fig ) . Staining with the anti-Iba1 antibody showed microgliosis mainly in the hippocampus , cerebral cortex and periaqueductal gray ( S3G-S3L Fig ) . Electron microscopy ( EM ) of Tg ( FFI ) brains detected neuronal ultrastructure abnormalities in several regions , including the neocortex , hippocampus , thalamus and cerebellum . These included autophagosomes , autophagolysosomes and increased amounts of lipofuscin ( Fig 10A–10D ) . The most marked finding , however , was alteration of the Golgi complex , whose cisternae appeared swollen and ‘swirled’ , often forming an onion-like structure ( Fig 10E–10G ) . Three-dimensional tomography confirmed the concentric isolated Golgi cisternae and showed invaginations of medial Golgi cisternae inside their lumen ( Fig 10H and 10I ) . These abnormalities were never seen in Tg ( WT-E1 ) and non-Tg controls . To see whether the Golgi abnormalities in Tg ( FFI ) neurons were associated with intracellular accumulation of mutant PrP , we examined primary CGNs by immuno-gold EM using a published procedure [14] . The majority of WT PrP in granule neurons from non-Tg/Prnp+/+ mice localized on the plasma membrane and in endosomes , with only a small fraction in the ER and Golgi ( Fig 11A and 11D ) . In contrast , D177N/M128 PrP localized mostly in the Golgi of Tg ( FFI ) neurons ( ~75% vs . ~2 . 5% in control cells ) , with far fewer molecules on the plasma membrane ( ~15% vs . ~85% in controls ) ( Fig 11B and 11D ) . The Golgi in these neurons were bigger than controls ( Fig 11E ) . These abnormalities in PrP distribution and intracellular organelle morphology were strikingly different from those of Tg ( CJD ) neurons , in which we found dramatic swelling of the ER cisternae , with ER retention of mutant PrP ( Fig 11C–11E ) [14] . Intracerebral inoculation of brain homogenates from FFI and CJD178 patients induced prion disease in experimental animals [22–24] , consistent with the contention that D178N PrP can spontaneously acquire an infectious structure . To test whether prion infectivity was generated de novo in the brains of Tg ( FFI ) and Tg ( CJD ) mice , we prepared different brain homogenates from Tg lines expressing 3F4-tagged or untagged mutant PrP at different levels , and co-expressing endogenous WT PrP or not ( Table 2 ) . The brain homogenates were inoculated intracerebrally in C57BL/6J mice , and in Tga20 mice that overexpress moPrP at 8X and are highly sensitive to prions [25] . Homogenates from Tg ( FFI-K5 ) and Tg ( CJD-A21 ) mice , expressing 3F4-tagged mutant PrPs , were also inoculated in Tg ( WT-E1+/+ ) mice overexpressing WT moPrP with the 3F4 epitope [13] , and in Tg ( CJD-G1+/+ ) mice , which express low levels of 3F4-tagged D177N/V128 PrP and do not spontaneously become ill [14] . The PrPs expressed by these two lines of mice should be particularly efficient for assaying infectivity because they contain either the 3F4 epitope or the 3F4 epitope and the D177N mutation . All transgenic recipient mice used in this study carried two disrupted Prnp alleles , so they did not synthesize endogenous PrP . Brain homogenates from non-Tg/Prnp+/+ mice served as negative controls . As positive controls , some host mice were inoculated with the mouse-adapted RML isolate of scrapie that had been previously passaged in WT or Tg ( WT-E1+/+ ) mice ( RML and RML3F4 , respectively ) [26] . All mice were observed weekly for the appearance of neurological signs . None of the animals inoculated with brain homogenates from Tg ( FFI ) and Tg ( CJD ) mice , or from negative control mice , developed neurological dysfunction , and all the animals either died from intercurrent illness or were euthanized near the end of their normal lifespan , approximately two years after inoculation ( Table 2 , lines 2–17 , 21–36 , 40–49 , 53–62 ) . None of the brains from inoculated C57BL/6J , Tg ( WT-E1+/+ ) or Tga20 host mice that were subjected to biochemical analysis contained PrP that was detergent-insoluble or that yielded typical or atypical ( i . e . 1E4- or SAF84-immunoreactive [27 , 28] ) PK-resistant fragments . Tg ( CJD-G1+/+ ) spontaneously accumulate small amounts of detergent-insoluble PrP in their brains [14] , but inoculation with Tg ( CJD ) or Tg ( FFI ) brain homogenates did not increase the amount . In contrast , all positive control mice inoculated with RML prions developed scrapie ( Table 2 , lines 18 , 19 , 37 , 38 , 50 , 51 ) , and their brains contained PrP that was resistant to high concentrations of PK . Thus , the brains of Tg ( FFI ) and Tg ( CJD ) mice did not contain prion infectivity detectable by bioassay . To test whether that the brains of Tg ( FFI ) and Tg ( CJD ) mice contained prions below the threshold of detection of the bioassay , we subjected the brain homogenates to serial protein misfolding cyclic amplification ( PMCA ) . This allows highly efficient prion replication in a test tube , and is able to amplify the equivalent of a single molecule of PrPSc [29] . Brain homogenates from Tg ( FFI ) and Tg ( CJD ) mice were subjected to serial PMCA with or without a RML seed . RML-seeded PMCA yielded forms of D177N/M128 and D177N/V128 PrPs that were highly PK-resistant ( hereafter referred to as D177N/M128RML and D177N/V128RML , respectively ) , while the unseeded reactions did not ( Fig 12A and 12B ) , indicating that the brains of Tg ( FFI ) and Tg ( CJD ) mice did not contain any spontaneously generated PrPSc . To test whether D177N/M128RML and D177N/V128RML corresponded to bona fide PrPSc , the PMCA reactions were intracerebrally inoculated in Tga20 mice . Four out of seven mice inoculated with D177N/M128RML , and five of nine inoculated with D177N/V128RML , had typical PK-resistant PrPSc in their brains ( Fig 12C and 12D ) , although at the time of death ( > 400 days post-inoculation: d . p . i . ) they did not have clinical signs of scrapie ( for comparison Tga20 mice inoculated with in vitro amplified RML developed scrapie at 78 ± 6 d . p . i . and died at 83 ± 6 d . p . i . ; mean ± SEM , n = 3 ) . Thus the in vitro converted mutant PrPs were able to propagate in vivo as authentic prions , but produced subclinical infections , most likely because of differences between the primary structure of the mutant PrPSc in the infecting inoculum and WT PrPC in the recipient mice [30 , 31] . As expected , the brains of Tga20 mice inoculated with the unseeded PMCA reactions did not contain any PK-resistant PrPSc ( Fig 12C and 12D ) .
The main EEG and sleep alterations of FFI patients [32] are also seen in Tg ( FFI ) /Prnp0/0 mice . Longitudinal 24-h monitoring and spectral EEG analysis show a marked reduction in sleep spindles in FFI patients [32] , and sleep spindle density is reduced by 75–80% in Tg ( FFI ) /Prnp0/0 mice compared to non-Tg controls . Although there is no change in the amount of NREM sleep , SWA during NREM sleep ( a measure of sleep drive and depth [17] ) is significantly reduced in Tg ( FFI ) /Prnp0/0 mice , like in FFI patients [32] . Tg ( FFI ) /Prnp0/0 mice also present a profound disruption of sleep continuity and organization . They have a larger number of transitions between the different behavioral states than non-Tg mice . In addition , in Tg ( FFI ) /Prnp0/0 mice approximately one fourth of REM sleep episodes starts directly from wakefulness , instead of being preceded by NREM sleep as normally occurs; this is reminiscent of the sudden-onset episodes of REM sleep that intrude into wakefulness in FFI patients [32 , 33] . The amount of REM sleep is also significantly decreased in Tg ( FFI ) /Prnp0/0 mice , as often happens in FFI patients at a late stage of disease [33] . Finally , Tg ( FFI ) /Prnp0/0 mice show an abnormal EEG pattern during REM sleep , with a significant reduction in theta activity . Cyclic organization of sleep and circadian motor rhythmicity are lost in FFI patients [32] , and we found that the circadian organization of sleep and motor activity was lost in Tg ( FFI ) /Prnp0/0 mice . Bursts of quasi-periodic sharp waves at 0 . 5–2 Hz ( similar to the periodic sharp-wave complexes in the EEG activity of CJD patients ) may appear in advanced stages of the long-evolution cases of FFI [32–34] . Bursts of high-voltage polyphasic complexes , similar to those described in Tg ( CJD ) mice [14] , were detected in the EEG of Tg ( FFI ) /Prnp0/0 mice . Interestingly , some sleep abnormalities were attenuated in Tg ( FFI ) /Prnp+/0 mice co-expressing endogenous WT PrP . For example the amount of REM sleep and circadian organization of sleep and motor rhythms are normal in these mice but they respond abnormally to sleep deprivation , not fully compensating the loss . Moreover , during the light phase , they show more sleep fragmentation than non-Tg/Prnp+/+ mice , and have less sleep spindle density and abnormal EEG activity , although less than Tg ( FFI ) /Prnp0/0 mice . PrPC has been assigned a role in promoting sleep continuity and circadian rhythmicity [35] . Our observation that re-introduction of one WT PrP allele confers some protection against the sleep and EEG changes induced by the D177N/M128 mutation suggests that functional loss of PrPC may contribute to sleep disruption in FFI mice . It would be interesting to see whether co-expression of WT PrP at a level comparable to that of transgenic mutant PrP completely reverses the sleep phenotype . The alterations in sleep and circadian rhythmicity in Tg ( FFI ) mice differ significantly from those in Tg ( CJD ) mice modeling CJD178 [14] . Whereas in Tg ( FFI ) mice the circadian organization of sleep and motor activity is lost , sleep is fragmented , and theta activity ( the EEG hallmark of rodent REM sleep ) is reduced , in Tg ( CJD ) mice the circadian organization of sleep , its continuity and its EEG patterns are not altered [14] . Thus Tg ( FFI ) and Tg ( CJD ) mice recapitulate specific phenotypic features of the corresponding human diseases . Ataxia and other motor symptoms , such as myoclonus , tremor , dysarthria and pyramidal impairment , are clinical features of FFI , which in some cases can be the earliest and most marked neurological signs [36–40] . We found that Tg ( FFI ) mice developed ataxia and sensorimotor deficits , such as inability to climb on a vertical grill , poor performance on a rotating rod , kyphosis and foot clasp reflex . The time of onset and rate of progression of these symptoms is directly correlated with transgene dosage . Tg ( FFI ) mice with mutant PrP expression below the level of endogenous PrP remain healthy throughout their lifetime . Mild neurological signs develop in ~60% of hemizygous Tg ( FFI-10 ) mice expressing mutant PrP at the endogenous level , and these animals live as long as nontransgenic controls , similar to knockin ki-3F4-FFI mice in which 3F4-tagged D177N/M128 PrP is under the control of the endogenous Prnp promoter [41] . In contrast , all Tg ( FFI ) mice expressing mutant PrP twice the endogenous level or more develop progressive and invariably fatal neurological disease . Transgenic mice expressing WT PrP up to seven times the endogenous level remain healthy during their lifetime [13 , 42] , and have normal sleep-wake behavior [14] , strongly indicating that the Tg ( FFI ) phenotype is not a mere consequence of overexpression , but is due to the D177N/M128 mutation . In contrast to the mitigating effect on sleep abnormalities , co-expression of WT PrP does not significantly modify the time of onset and progression of motor dysfunction or prolong survival of Tg ( FFI ) mice , consistent with the dominant mode of inheritance of FFI and with similar observations in other mutant PrP mice [14 , 19 , 43] . Thus , WT PrP influences only some aspects of the Tg ( FFI ) phenotype , perhaps those that are more dependent on a physiological function of PrP in neuronal excitability [44] . Disturbances of attention and memory , difficulties with the temporal ordering of events and spatial disorientation are early signs of FFI , which usually appear before ataxia and other motor deficits [32 , 45] . Tg ( FFI ) mice have significant alterations in recognition and spatial working memory , detectable in the object recognition test and the eight-arm radial maze before the onset of motor dysfunction . Thus salient motor and cognitive aspects of the clinical picture of FFI are recapitulated in transgenic mice . Tg ( FFI ) mice also show brain abnormalities reminiscent of human FFI . Thalamic degeneration is the most marked neuropathological change in FFI patients , who frequently also have degenerated cerebella [36 , 40 , 46] , and reduced thalamic and cerebellar volumes were detected by MRI in Tg ( FFI ) mice in the advanced stage of disease . Although PrPSc accumulates to lower levels in FFI than in other human prion diseases , widespread protease-resistant PrP deposits can be detected by immunohistochemistry in the CNS of FFI patients , especially in cases of long duration [47] . These include fine-granular synaptic-type and focal PrP deposits , patchy and strip-like immunoreactive profiles , as well as intravacuolar and cytoplasmic PrP accumulations in neurons [40] . Synaptic-type , dot-like and small round PrP profiles , as well as immunoreactive fiber tracts and intraneuronal PrP deposits , reminiscent of those in FFI patients , are detected in the brains of Tg ( FFI ) mice . Moderate gliosis is seen in several brain regions of Tg ( FFI ) mice , including the external layers of the cerebral cortex and the periaqueductal gray , similar to findings in FFI patients where gliosis has a bilaminar accentuation in the cerebral cortex and is often detected in the periaqueductal gray [40 , 47] . Mutant PrP in the brains of Tg ( FFI ) and Tg ( CJD ) mice displays several biochemical characteristics reminiscent of PrPSc , including insolubility in non-denaturing detergents , low PK resistance and reactivity with PrPSc-directed antibodies [14 , 48 , 49 and this study] , raising the possibility that the protein may have spontaneously acquired an infectious conformation . However , we found that brain homogenates from Tg ( FFI ) and Tg ( CJD ) mice had no detectable infectivity when inoculated into different transgenic and nontransgenic hosts . We also found no amplification of PrPSc in unseeded PMCA reactions , ruling out the possibility that the inocula contained prions below the threshold of detection by our bioassay . Thus mutant PrP in the brains of Tg ( FFI ) and Tg ( CJD ) mice is misfolded and pathogenic but unable to propagate its abnormal conformation , and is therefore fundamentally different from PrPSc . However , it is not intrinsically refractory to PrPSc conversion since it acquires an infectious conformation when subjected to PMCA in the presence of a RML seed . Our results are different from those of Jackson et al . , who reported the emergence of spontaneous prion infectivity in knockin ki-3F4-FFI mice expressing 3F4-tagged D177N/M128 PrP from the endogenous Prnp locus [41] . Ki-3F4-FFI mice did not die prematurely , but developed behavioral abnormalities in old age , similar to our hemizygous Tg ( FFI-10 ) mice expressing mutant PrP at the endogenous level . Intracerebral inoculation of brain homogenates from sick ki-3F4-FFI mice induced neurological disease in ki-3F4-WT mice expressing 3F4-tagged WT PrP , and in Tga20 mice , but not in nontransgenic mice [41] , arguing that homology between mutant PrP and the recipient’s PrPC at residues 108 and 111 ( which constitute the 3F4 epitope ) , or overexpression of untagged PrP by the recipient mice , is required for disease transmission . In contrast with this , we found that the Tg ( FFI ) disease could not be transmitted to other mice , despite homology between mutant and the recipient’s PrP , and/or PrP overexpression in the recipient mice . The reason for this difference is not clear . It was suggested that mutant PrP needs to be targeted to the Prnp locus to generate a transmissible agent spontaneously [41] . However , prions emerged spontaneously even in mice with randomly integrated transgenes , including those constructed using the half-genomic Prnp vector used in our study [21 , 50–53] . In addition , a gene targeting approach identical to the one used by Jackson et al . was employed to generate P101L mice expressing the mouse homolog of the P102L mutation linked to GSS [54] , and these mice did not develop a transmissible prion disease [54] . Thus , replacing the endogenous PrP coding sequence is neither necessary nor sufficient for de novo prion generation . Ki-3F4-FFI mice have a mixed 129/Ola X C57BL/6N genetic background , whereas our Tg ( FFI ) mice are C57BL/6J x CBA/J hybrids backcrossed with C57BL/6J mice . It is possible that the genetic makeup of ki-3F4-FFI mice favors generation of prion infectivity , for example because co-factors that might promote PrPSc formation [55 , 56] are selectively expressed or enriched in these animals . However , prion infectivity developed de novo even in Tg PrP mice with C57BL/6J X 129S5 [50] , C57BL/6 X FVB [52] , C57BL/6J X CBA/J X 129/Ola [21] and FVB [51 , 53] backgrounds , suggesting that a specific allelic composition is not required . Mutant PrP in the brains of ki-3F4-FFI mice does not show the typical biochemical attributes of PrPSc , suggesting that ki-3F4-FFI mice bear an unconventional prion strain [41] . However , small amounts of protease resistant PrP were detected when 20 mg of total brain homogenates from two-year-old ki-3F4-FFI mice was PK digested and concentrated by trichloroacetic acid precipitation [41] . It would be interesting to see whether this form of the protein can be amplified by PMCA . It would also be interesting to compare the conformations of D177N/M128 PrP molecules extracted from the brains of ki-3F4-FFI and our Tg ( FFI ) mice , because this may shed light on the structural features that enable mutant PrP to self-replicate . In this regard , we have found that infectious and non-infectious forms of PG14 PrP are structurally related but differ in their oligomeric state and degree of PK resistance [26 , 48 , 57] . Different prion diseases are thought to be enciphered in distinctive conformations of the PrPSc molecule ( prion strains ) [24 , 58] . Our observation that Tg ( FFI ) and Tg ( CJD ) mice recapitulate specific features of the respective human disorders without developing prion infectivity—and the same holds true for Tg ( PG14 ) mice and other mouse models of GSS [20 , 26 , 59 and J . Mastrianni , personal communication]—suggests that the disease-encoding properties of mutant PrP are enciphered in misfolded conformations of the protein that are toxic but not infectious [60 , 61] . We recently obtained evidence that the mutant PrPs extracted from the brains of Tg ( FFI ) , Tg ( CJD ) and Tg ( PG14 ) mice have different structures , indicating that they carry enough conformational diversity to encode different diseases [49] . Our observation that the pathogenicity of misfolded PrP does not depend on its ability to self-replicate has implications for other neurodegenerative diseases associated with protein misfolding . There is evidence that aggregated proteins such as amyloid-β in AD , tau in the tauopathies , and α-synuclein in PD , are able to spread in a prion-like manner when inoculated into the mouse brain [62] . However , the mechanism linking spread of protein misfolding and neurodegeneration in these diseases is not clear [63] . Our results are consistent with a dissociation between the toxic and propagating PrP species [60 , 61] , and suggest that there may be a separation between spread of misfolded proteins and neurotoxicity also in other neurodegenerative diseases . We have reported that mouse PrP molecules carrying the D177N mutation are delayed in their biosynthetic maturation and are partially retained in transport organelles of the secretory pathway [14 , 64–66] . In the present study we found that D177N/M128 PrP accumulates preferentially in the Golgi of Tg ( FFI ) neurons , and this is associated with enlargement of the Golgi . In contrast , D177N/V128 PrP is mostly found in the ER of Tg ( CJD ) neurons , and this organelle appears swollen and electrondense [14] . Thus the two polymorphic variants tend to accumulate in different intracellular compartments , causing different morphological alterations of transport organelles . The intracellular accumulation of D177N/M128 and D177N/V128 PrPs also differed in transfected cells [65 , 67] , and may reflect the way these mutants acquire abnormal conformations and aggregate during secretory transport [68] . There is evidence , in fact , that the M/V 129 polymorphism influences the kinetics of misfolding and oligomerization of D178N PrP [11] , and oligomerization and intracellular retention of this mutant are closely related [69] . We previously found that ER retention of PG14 and D177N/V128 PrPs impairs the transport of VGCCs to presynaptic terminals of CGNs , due to a physical interaction of PrP with the auxiliary α2δ-1 subunit of the channel [15] . Since artificial targeting of PrP to the Golgi results in intracellular retention of α2δ-1 [15] , D177N/M128 may also impair VGCC transport and function in CGNs , contributing to the motor dysfunction of Tg ( FFI ) , as in Tg ( PG14 ) and Tg ( CJD ) mice . However , there could be other PrPC-interacting proteins whose cellular trafficking and synaptic targeting may be affected differently by the different mutants , potentially triggering specific neurotoxic effects [70] . PrPC interacts physically with the NR1 and NR2D subunits of N-methyl-D-aspartate ( NMDA ) and the GluA1 and GluA2 subunits of α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid ( AMPA ) receptors , and these interactions are important for normal neuronal physiology and survival [71–73] . Interestingly , the assembly and trafficking of these receptors are finely tuned in the ER and Golgi [74 , 75] . Our preliminary observations indicate that mutant PrPs that accumulate in different intracellular organelles affect NMDA and AMPA receptor trafficking in different ways . Moreover , different mutants interact differently with receptor subunit isoforms expressed in functionally distinct neurons of the brain [73] . Thus , different mutant PrPs may have different effects on the function and survival of different neurons—hence on the clinical presentation of disease—depending on where in the secretory pathway they preferentially localize , and how this interferes with the transport of the molecules they interact with [70] . In summary , we have generated transgenic mice that model essential aspects of FFI , and differ from analogous mice expressing the CJD178 mutation . Disease-specific features are seen in independently generated Tg lines with different copies of integrated transgene and PrP expression levels , strongly indicating that they are encoded by mutant PrP , rather than non-specific effects of random transgenesis . Tg ( FFI ) mice may be useful for investigating the pathophysiology of sleep in FFI and for testing potential therapies for this devastating disorder . Comparative studies of Tg ( FFI ) and Tg ( CJD ) mice may provide important information on the molecular mechanisms responsible for the phenotypic heterogeneity associated with the polymorphic variants of the PrP D178N mutation .
The cDNAs encoding mouse PrP derived from the Prnpa allele , containing the D177N/M128 substitution with or without the 3F4 epitope tag , were ligated into the MoPrP . Xho vector , which contains a 12 kb fragment of Prnp , including the promoter and intron 1 , and drives the expression of transgenic PrP in a tissue pattern similar to that of the endogenous protein [14 , 76] . Recombinant plasmids were selected by PCR screening and restriction analysis , and their identity confirmed by sequencing the entire coding region [13] . The transgene was excised by NotI digestion and injected into the pronuclei of fertilized eggs from an F2 cross of C57BL/6J x CBA/J F1 parental mice . Transgenic founders were bred with an inbred colony of Zürich I Prnp0/0 mice [77] with a pure C57BL/6J background ( C57BL/6J/Prnp0/0; European Mouse Mutant Archive , Monterotondo , Rome , Italy; EM:01723 ) . The status of the Prnp gene and the presence of the transgene were determined by PCR , and the zygosity of the transgene by RT-PCR [13] . Production of transgenic Tg ( WT-E1+/- ) mice , expressing wild-type PrP tagged with an epitope for monoclonal antibody 3F4 at approximately 2X , and Tg ( CJD-A21+/- ) and Tg ( CJD-G1+/+ ) mice expressing 3F4-tagged D177N/V128 PrP at ~1X and ~0 . 3X respectively , has already been reported [13 , 14] . Tg ( CJD-66+/- ) and Tg ( CJD-39+/- ) mice , which express untagged D177N/V128 PrP at ~2X and ~4X respectively , were generated as described above . They develop a CJD-like neurological syndrome like Tg ( CJD-A21 ) mice [14] , which will be described in details elsewhere . All transgenic lines used in this study were backcrossed for at least ten generations with C57BL/6J/Prnp0/0 mice , with the exception of Tga20 ( EM:00181 ) [25] , Tg ( CJD-G1 ) ( EM:06144 ) and Tg ( WT-E1+/+ ) [14] mice which were maintained as inbred hybrid colonies . For some experiments the Prnp allele was re-introduced by breeding transgenic mice with C57BL/6J ( Harlan Laboratories ) . Procedures involving animals and their care were conducted in conformity with the institutional guidelines at the IRCCS—Mario Negri Institute for Pharmacological Research in compliance with national ( D . lgs 26/2014; Authorization n . 19/2008-A issued March 6 , 2008 by Ministry of Health ) and international laws and policies ( EEC Council Directive 2010/63/UE; the NIH Guide for the Care and Use of Laboratory Animals , 2011 edition ) . They were reviewed and approved by the Mario Negri Institute Animal Care and Use Committee that includes ad hoc members for ethical issues ( 18/01-D and 18/02-D ) , and by the Italian Ministry of Health ( Decreto nr 62/2012-B and 63/2012-B ) . Animal facilities meet international standards and are regularly checked by a certified veterinarian who is responsible for health monitoring , animal welfare supervision , experimental protocols and review of procedures . Assays of detergent insolubility and proteinase K resistance were done as described [13] . Western blots were developed with monoclonal anti-PrP antibodies 12B2 ( Central Veterinary Institute , Wageningen , NL ) , 1E4 ( Cell Sciences ) , or SAF84 ( Spi Bio ) . We investigated EEG and sleep patterns in eight non-Tg/Prnp+/+ , ten non-Tg/Prnp0/0 , nine Tg ( FFI-26+/- ) /Prnp0/0 , and eight Tg ( FFI-26+/- ) /Prnp+/0 mice aged between 332 and 468 days of age . Mice were anesthetized and instrumented for chronic EEG recording according to standard techniques [78] . They were individually housed in standard cages with food and water ad libitum , and allowed at least ten days to recover from surgery and adapt to the recording conditions . Cages were kept in sound-attenuated rooms at a constant temperature ( 26 ± 1°C ) , with a 12/12-hour light-dark cycle . Gross body activity was detected using an infrared sensor housed in an observation unit that also contained a camera ( BioBserve GmbH , Bonn , Germany ) , allowing undisturbed monitoring of the animals’ behavior . Movements detected by the infrared sensor were converted to a voltage output . The conditioned EEG signal and the voltage output from the infrared sensor were digitized and collected using custom software ( M . R . Opp , University of Michigan ) . For investigations in undisturbed conditions , EEG signals and gross body activity were recorded for 24 h ( starting at the beginning of the light phase ) and used for polygraphic determination of vigilance . The animals were not handled starting from 48 h before the recording session . For investigations of the effects of sleep deprivation , mice were sleep-deprived by gentle handling for the first six hours of the light phase , then allowed to behave and sleep freely for the next 18 hours ( i . e . for the second six hours of the light phase followed by the12 hours of the dark phase ) . Two mice of different genotypes were always randomly matched and recorded simultaneously . The order of recording of mice of the different lines was also randomized . Postacquisition determination of vigilance was done according to standard criteria [78] . An investigator ( F . D . G . ) blind to the strain visually scored 12-s epochs . EEG power densities were obtained for each animal and each behavioral state by Fourier transform for each artifact-free 12-s scoring epoch for the frequency range 0 . 5–20 Hz . Values in the 0 . 5–4 . 0 Hz ( delta ) frequency range were collapsed and integrated for 12-s epochs , and used as measures of SWA during NREM sleep . EEG recordings during the light portion of the light-dark phase were also band-pass filtered ( 10–13 Hz , 4th order Chebyschev type II filter ) , and NREM sleep spindles were visually identified . Spindle density was obtained ( number of spindles/hour of NREM sleep ) . Statistical analysis was done as described in the Results section or in the legends to tables and figures . Mice were observed weekly for signs of neurological dysfunction , according to a set of objective criteria [13] . Onset of disease was scored as the time at which at least two neurological signs were observed , out of foot-clasp reflex , kyphosis , unbalanced body posture , inability to walk on a horizontal metal grid , and to remain on a vertical grid for at least 30 s . The accelerating Rotarod 7650 model ( Ugo Basile ) was used: mice were first trained three times the week before official testing . They were positioned on the rotating bar and allowed to become acquainted with the environment for 30 s . The rod motor was started at an initial setting of 7 r . p . m . and accelerated to 40 r . p . m . at a constant rate of 0 . 11 r . p . m . /s for a maximum of 300 s . Performance was scored as latency to fall , in seconds . Animals were given three trials , and the average was used for statistical analysis . Spatial working memory was measured using an eight-arm radial maze made of grey plastic with a Plexiglas lid . The arms radiating from an octagonal central arena with a diameter of 12 cm , were 30 cm long , 5 cm wide and 4 cm high . Several extra-maze visual cues surrounded the apparatus . Starting one week before testing , the mice were water-deprived by allowing them water for only one hour a day . One day before starting the task schedule , a habituation trial was run . The mice were placed in the center of the maze and left free to explore the environment for 5 minutes . The next day the arms of the radial maze were baited with 50 μl of water . Animals were placed in the center of the maze and the arm-entry sequence was recorded . The task ended once all eight arms of the maze had been visited or after a maximum of 16 trials , whichever came first . Repeated entry into an arm that had already been visited constituted an error . The number of errors and the latency to complete the test were recorded manually by an operator ( I . B . ) blind to the experimental groups . Animals were tested for 16 consecutive days . Mice were tested in an open-square grey arena ( 40 x 40 cm ) , 30 cm high , with the floor divided into 25 squares by black lines . The following objects were used: a black plastic cylinder ( 4 x 5 cm ) , a glass vial with a white cap ( 3 x 6 cm ) and a metal cube ( 3 x 5 cm ) . The task started with a habituation trial during which the animals were placed in the empty arena for 5 minutes and their movements were recorded as the number of line-crossings . The next day , mice were placed in the same arena containing two identical objects ( familiarization phase ) . Exploration was recorded in a 10-minute trial by an investigator ( C . B . ) blinded to the experimental group . Sniffing , touching and stretching the head toward the object at a distance of not more than 2 cm were scored as object investigation . Twenty-four hours later ( test phase ) mice were placed in the arena containing two objects: one identical to one of the objects presented during the familiarization phase ( familiar object ) , and a new , different one ( novel object ) , and the time spent exploring the two objects was recorded for 10 min . Memory was expressed as a discrimination index , i . e . the time spent exploring the novel object minus the time spent exploring the familiar object , divided by the total time spent exploring both objects; the higher the discrimination index , the better the performance . Animals were anesthetized with 1% isoflurane in a 30:70% O2:N2O gas mixture and imaged in a horizontal bore 7-Tesla USR preclinical MRI system ( BioSpec 70/30 , Bruker BioSpin , Germany ) with a shielded gradient insert ( BGA 12 , 400 mT/m; rise time , 110 us ) . A 72-mm birdcage resonator for RF transmission , and a 10-mm diameter single-loop receiver coil were used to receive the signal . 3D T2-weighted anatomical images of the mouse brain were acquired with the following parameters: TR 2500 ms , TE 50 ms , RARE factor 16 , FOV 3 x 1 . 5 x 1 . 5 cm , Matrix 256 x 102 x 102 , voxel 0 . 147 x 0 . 117 x 0 . 147 . The scan time was approximately 25 min . The volumes of the whole brain and individual brain areas ( frontal cortex , hippocampus , thalamus , striatum and cerebellum ) were quantified manually using Fiji software [79] , after a rigid body registration ( 6 dof ) to a reference image to avoid bias due to bad head positioning . Mice were euthanized by CO2 inhalation , brains were removed and fixed in Alcolin ( Diapath ) or Carnoy’s fixative ( ethanol , chloroform , acetic acid , 6:3:1 ) , dehydrated in graded ethanol solutions , cleared in xylene , and embedded in paraffin . Serial sections ( 8 mm thick ) were cut and stained with hematoxylin and eosin , Nissl , or thioflavin S . For PrP immunohistochemistry , sections were incubated with PK ( 2 μg/ml in H2O ) for 1 h at room temperature , and exposed to guanidine thyocianate ( 3M in H2O ) for 30 min [80] . PK-resistant PrP was detected with monoclonal antibody 12B2 ( 1:2000 ) , using the ARK kit ( Dako ) , with 3 , 3’ diaminobenzidine ( DAB ) as chromogen . For glial fibrillary acidic protein ( GFAP ) and ionized calcium binding adapter molecule 1 ( Iba1 ) immunohistochemistry , mice were deeply anesthetized by intreperitoneal injection of 100 mg/kg ketamine hydrochloride and 1 mg/kg medetomidine hydrochloride ( Alcyon ) , and perfused through the ascending aorta with phosphate buffered saline ( PBS , 0 . 05 M; pH 7 . 4 ) followed by 4% paraformaldehyde ( PFA ) in PBS . Brains were removed , post-fixed , cryoprotected and frozen at -80°C . Sections were cut using a Leica cryostat and incubated for 1 h at RT with 10% normal goat serum ( NGS ) , 0 . 3% Triton X-100 in PBS 0 . 1M , pH 7 . 4 , then overnight at 4°C with mouse monoclonal anti-GFAP antibody ( Millipore , 1:2500 ) or rabbit polyclonal anti-Iba1 ( Wako Reagent , 1:1000 ) , followed by visualization with the Vectastain ABC kit ( Vector ) , using DAB as chromogen . One 503-day-old Tg ( WT-E1+/- ) /Prnp0/0 mouse , one 269-day-old Tg ( WT-E1+/+ ) /Prnp0/0 mouse , two non-Tg/Prnp0/0 mice aged 208 and 374 days , four non-Tg/Prnp+/+ mice 152 , 268 , 280 and 392 days old , one 302 and two 367 days old Tg ( FFI-26+/- ) /Prnp0/0 mice , one 444 days old Tg ( FFI-26+/- ) /Prnp+/0 and two Tg ( FFI-10+/- ) /Prnp0/0 mice aged 331 and 379 days were analyzed by electron microscopy . Mice were deeply anesthetized and perfused through the ascending aorta with phosphate buffered saline ( PBS , 0 . 1 M; pH 7 . 4 ) followed by 4% paraformaldehyde ( PFA ) and 2 . 5% glutaraldehyde in PBS . The brain was excised , cut along the sagittal plane with a razor blade , and postfixed in 3% glutaraldehyde in PBS then for 2 h in OsO4 . After dehydration in graded series of ethanol , tissue samples were cleared in propylene oxide , embedded in epoxy medium ( Epon 812 Fluka ) and polymerized at 60°C for 72 h . From each sample , one semithin section ( 1 μm ) was cut with a Leica EM UC6 ultramicrotome and mounted on glass slides for light microscopic inspection to identify the area of interest . Ultrathin sections ( 70 nm thick ) were obtained , counterstained with uranyl acetate and lead citrate , and examined with an Energy Filter Transmission Electron Microscope ( EFTEM , ZEISS LIBRA® 120 ) equipped with a YAG scintillator slow-scan CCD camera . Lipofuscin granules and cytoplasm of neurons were manually outlined , and areas were calculated with image analysis software ( iTem , Olympus ) . The sum of the areas occupied by lipofuscin in each neuron was expressed as a percentage of the cytoplasmic area . Electron tomography and three-dimensional reconstruction was done as described [81] . Immuno-electron microscopy of PrP in cultured cerebellar granule neurons , quantification of gold particles in the different compartments of the secretory pathway , and analysis of total cell , ER and Golgi volumes , were as described [14] . The detailed protocol for serial PMCA has been published elsewhere [82] . Briefly , 50 μl of 10% brain homogenate were loaded into 0 . 2-ml PCR tubes and positioned on an adaptor on the plate holder of a S-700MPX sonicator ( QSonica , Newtown , CT , USA ) . Each PMCA cycle consisted of 30 min incubation at 37°C followed by a 20 s sonication pulse at a potency of 70–90 . Samples were incubated without shaking immersed in the water of the sonicator bath . After a round of 48 cycles , a 10 μl portion of the amplified material was diluted into 90 μl of brain homogenate and a new round of 48 PMCA cycles was done . This procedure was repeated 10 times . For seeded PMCA , RML-derived , in vitro amplified PrPSc was added to the brain homogenates before PMCA . Samples were digested with 80 μg/ml of PK for 1h at 42ºC and analyzed by Western blot with anti-PrP antibody SAF84 . To produce enough material for transmission studies , and to rule out the possibility of residual RML in the seeded PMCA inocula , the reactions were run for seven additional rounds ( final dilution 10-20 ) [29] . Ten-percent ( w/v ) homogenates of mouse brain were prepared in PBS and cleared by centrifugation at 900 x g for 5 min . Serial PMCA reactions were diluted 1:10 in PBS before inoculation . 25 μl of the cleared homogenate , or diluted PMCA reactions , was injected intracerebrally into the right parietal lobe of recipient mice using a 25-gauge needle . | Genetic prion diseases are degenerative brain disorders caused by mutations in the gene encoding the prion protein ( PrP ) . Different PrP mutations cause different diseases , including Creutzfeldt-Jakob disease ( CJD ) and fatal familial insomnia ( FFI ) . The reason for this variability is not known , but assembly of the mutant PrPs into distinct aggregates that spread in the brain by promoting PrP aggregation may contribute to the disease phenotype . We previously generated transgenic mice modeling genetic CJD , clinically identified by dementia and motor abnormalities . We have now generated transgenic mice carrying the PrP mutation associated with FFI , and found that they develop severe sleep abnormalities and other key features of the human disorder . Thus , transgenic mice recapitulate the phenotypic differences seen in humans . The mutant PrPs in FFI and CJD mice are aggregated but unable to promote PrP aggregation . They accumulate in different intracellular compartments and cause distinct morphological abnormalities of transport organelles . These results indicate that mutant PrP has disease-encoding properties that are independent of its ability to self-propagate , and suggest that the phenotypic heterogeneity may be due to different effects of aggregated PrP on intracellular transport . Our study provides new insights into the mechanisms of selective neuronal dysfunction due to protein aggregation . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [] | 2015 | Transgenic Fatal Familial Insomnia Mice Indicate Prion Infectivity-Independent Mechanisms of Pathogenesis and Phenotypic Expression of Disease |
In the conserved autophagy pathway , the double-membrane autophagosome ( AP ) engulfs cellular components to be delivered for degradation in the lysosome . While only sealed AP can productively fuse with the lysosome , the molecular mechanism of AP closure is currently unknown . Rab GTPases , which regulate all intracellular trafficking pathways in eukaryotes , also regulate autophagy . Rabs function in GTPase modules together with their activators and downstream effectors . In yeast , an autophagy-specific Ypt1 GTPase module , together with a set of autophagy-related proteins ( Atgs ) and a phosphatidylinositol-3-phosphate ( PI3P ) kinase , regulates AP formation . Fusion of APs and endosomes with the vacuole ( the yeast lysosome ) requires the Ypt7 GTPase module . We have previously shown that the Rab5-related Vps21 , within its endocytic GTPase module , regulates autophagy . However , it was not clear which autophagy step it regulates . Here , we show that this module , which includes the Vps9 activator , the Rab5-related Vps21 , the CORVET tethering complex , and the Pep12 SNARE , functions after AP expansion and before AP closure . Whereas APs are not formed in mutant cells depleted for Atgs , sealed APs accumulate in cells depleted for the Ypt7 GTPase module members . Importantly , depletion of individual members of the Vps21 module results in a novel phenotype: accumulation of unsealed APs . In addition , we show that Vps21-regulated AP closure precedes another AP maturation step , the previously reported PI3P phosphatase-dependent Atg dissociation . Our results delineate three successive steps in the autophagy pathway regulated by Rabs , Ypt1 , Vps21 and Ypt7 , and provide the first insight into the upstream regulation of AP closure .
In autophagy , parts of the cytoplasm , including organelles , are engulfed by the phagophore ( or isolation ) membrane , which expands and closes to form the double-membrane autophagosomes ( APs ) . APs then fuse with the cellular degradative compartment termed lysosome , and the degradation products are recycled back to the cytoplasm . This process is important for the ability of cells to respond to stress , and is involved in the etiology of multiple human diseases [1 , 2] . A set of core autophagy-related proteins , Atgs , were identified in yeast and shown to be conserved from yeast to human cells [3] . These Atgs , together with membrane , assemble to form the pre-autophagosomal structure ( PAS ) , which is required for the formation of the isolation membrane and its expansion . However , currently very little is known about mechanisms that specifically regulate closure of APs , which is required for their successful fusion with the lysosome and the resulting delivery of single-membrane surrounded cargo for degradation [4 , 5] . A family of conserved GTPases , eleven Ypts ( includes Vps21 and Sec4 ) in yeast and seventy Rabs in mammalian cells , regulates all membrane-associated intracellular trafficking pathways . These GTPases are activated by their cognate guanine-nucleotide exchange factors , GEFs , and when on membranes in the GTP-bound state , they recruit their downstream effectors . Rab effectors include all the known membrane trafficking machinery components , such as vesicle coats , cytoskeletal motors , tethering factors and SNAREs [6] . Recently , Rab GTPases also emerged as regulators of autophagy [7] . In yeast , Ypt1 is required for ER-to-Golgi transport in the secretory pathway , and is also essential for PAS formation . Interestingly , Ypt1 regulates these two very different processes in the context of two distinct modules , namely , using different GEFs and effectors [8] . Ypt7 , which is required in the endocytic pathway for endosome fusion with the vacuole ( the yeast lysosome ) [9 , 10] , is also essential in autophagy for fusion of APs with the vacuole . Ypt7 , unlike Ypt1 , functions in endocytosis and autophagy with the same module of GEF and effectors [11–13] . The yeast proteome contains three Rab5-related proteins: Vps21 ( Ypt51 ) , Ypt52 and Ypt53 [14] . A Vps21 GTPase module regulates a step in the endocytic pathway that precedes the one controlled by the Ypt7 GTPase module [9] . A role for Rab5 in autophagy in yeast was originally suggested based on selective and general autophagy defects of vps21Δ ypt52Δ double deletion mutant cells ( but not of vps21Δ single deletion mutant cells ) [15] . More recently , based on selective and general autophagy defects of vps21Δ ( single ) deletion cells , we have shown that Vps21 regulates autophagy in the context of its endocytic GTPase module . Specifically , depletion of Vps21 , its GEF , or its effectors , results in accumulation of APs [16] . The human Rab1 , Rab5 and Rab7 GTPases , homologs of the yeast Ypt1 , Vps21 and Ypt7 , respectively , were also implicated in autophagy [17] . In addition to membrane sealing , AP maturation also includes the removal of certain Atgs [13] . Previously , it was shown that the PI3P phosphatase Ymr1 plays a role in Atg dissociation from APs . Specifically , depletion of Ymr1 results in the accumulation of sealed APs decorated with Atgs [18] . Thus , while PI3P generation by a PI3P kinase is required for AP formation [19 , 20] , PI3P removal by a PI3P phosphatase from sealed APs is required for Atg dissociation , which is in turn required for AP fusion with the vacuole . While we have shown that the Vps21 GTPase module plays a role in autophagy in yeast , it was not clear which step of the autophagy pathway this module regulates . Here , we show that the Vps21 GTPase module plays a role before the elusive step of AP closure . Moreover , we show that the Vps21-regulated step precedes Ymr1-mediated Atg dissociation in AP maturation . Finally , using double mutant analyses , we show that Vps21 functions between Ypt1-mediated AP formation and Ypt7-dependent AP fusion with the vacuole .
In yeast , autophagy can be induced either by nitrogen starvation or by addition of rapamycin [21] . We and others have previously shown that an autophagy-specific mutation in Ypt1 or depletion of its autophagy-specific GEF subunit , Trs85 , results in a defect in PAS formation under normal growth conditions and when autophagy is induced by nitrogen starvation [8 , 22–24] . In contrast , depletion of the Vps21 GTPase module components causes accumulation of APs under nitrogen starvation , and in most cells , AP clusters are seen near the vacuolar membranes [16] . Vps21 together with its paralog Ypt52 was proposed to have a role in autophagy based on autophagy defects in the double-deletion mutant cells when general autophagy was induced by rapamycin [15] . To check if Ypt1 and Vps21 GTPases function in the same pathway , we used double-mutant epistasis analysis . In this analysis , we took advantage of the different phenotypes of mutations in YPT1 and VPS21 to determine which phenotype masks the other . If Ypt1 functions upstream of Vps21 in autophagy , the phenotype of ypt1-1 vps21Δ double mutant cells should be similar to that of ypt1-1 and not vps21Δ single mutant cells . PAS formation was determined by co-localization of two PAS/AP markers , Atg11 and Atg8 , tagged with GFP and mCherry , respectively , during normal growth , and when autophagy is induced by nitrogen starvation or rapamycin , using live-cell fluorescence microscopy . In wild type cells , Atg8 and Atg11 co-localized to a single dot per cell ( 60–70% of the Atg8 dots ) that represents PAS or AP . In ypt1-1 mutant cells , both Atgs appeared as multiple dots per cell , and most of these dots did not co-localize ( 65–70% ) . In contrast , in most vps21Δ mutant cells , Atg11 and Atg8 co-localized on APs ( 65–75% ) and most cells accumulate AP clusters ( seen as crescents near the vacuole ) when autophagy was induced . The ypt1-1 vps21Δ double mutant cells exhibited a phenotype similar to that of ypt1-1 , namely , Atg8 and Atg11 appeared as multiple dots that did not co-localize ( 65–70% ) . Moreover , whereas ~65–75% of the vps21Δ mutant cells accumulated AP clusters when autophagy was induced by nitrogen starvation or rapamycin , no AP clusters were observed in ypt1-1 mutant and in ypt1-1 vps21Δ double mutant cells ( Fig 1 ) . The mis-localization of Atg8 and Atg11 in both ypt1-1 and ypt1-1 vps21Δ mutant cells was not caused by a significant decrease in protein levels ( S1C and S1D Fig ) . These results indicate that PAS assembly is defective in ypt1-1 vps21Δ double mutant cells . Thus , Ypt1 functions upstream of Vps21 in both selective ( normal growth ) and non-selective ( nitrogen starvation or rapamycin ) autophagy . The idea that the Vps21 GTPase module functions in a late step of autophagy is supported by accumulation of AP clusters in mutant cells depleted for components of this module . Additional support comes from analysis of Atg8 lipidation in these mutant cells . Atg8 lipidation is required for the attachment of Atg8 to membranes , and therefore for both AP formation and expansion [3] . The level of lipidated Atg8 , Atg8-PE , was determined using immunoblot analysis in wild type , vps21Δ and vps9Δ mutant cells under nitrogen starvation . This analysis shows that similar levels of Atg8-PE are present in wild type , vps21Δ and vps9Δ mutant cells ( S3 Fig ) . Together , the accumulation of AP clusters and the Atg8-PE level in vps21Δ and vps9Δ mutant cells indicate that the Vps21 GTPase module is not required for AP formation or expansion , but in a successive step . Ypt7 is required for AP fusion with the vacuole [11] . Because the APs that accumulate in ypt7Δ mutant cells are sealed , the enclosed cargo is protected from degradation , as determined by an immunoblot analysis in a protease-protection assay ( see Materials and Methods ) . Using this assay , autophagic cargo accumulating in atg mutant cells , which are defective in PAS and APs assembly , is sensitive to degradation [25] . Until this study , there was no mutant known to accumulate APs with cargo sensitive to degradation . This protease protection assay was used to determine whether APs that accumulate in vps21Δ mutant cells are sealed . Protection of two autophagy cargos was tested: prApe1 and GFP-Atg8 . When not enclosed inside membranes , these cargos can be cleaved to mApe1 and GFP , respectively , by addition of proteinase K ( PK ) to the cell fraction ( P5 ) that contains membrane-bound compartments . In atg1Δ mutant cells , in which incomplete PAS can be formed but cannot expand [26–28] , the cargos were not protected , and the cleaved products are seen after addition of PK . In contrast , in ypt7Δ mutant cells , which accumulate APs [11] , ~60% of the prApe1 and 45% of the GFP-Atg8 were protected from the protease . Importantly , the fact that the two cargos can be cleaved upon solubilization of the membranes by a detergent ( Triton X-100 , TX ) when prepared from ypt7Δ mutant cells , shows that they were protected by sealed membranes . Notably , while APs accumulate in vps21Δ mutant cells , prApe1 and GFP-Atg8 were not protected from the protease and are cleaved ( Fig 2A and 2B ) . This phenotype is different from that exhibited by ypt7Δ mutant cells , and suggests that APs accumulating in vps21Δ mutant cells are not sealed . Moreover , while APs accumulate in vps21Δ ypt7Δ double mutant cells ( see below ) , the two autophagy cargos were also not protected from the protease ( Fig 2A and 2B ) . Thus , as in vps21Δ mutant cells , in the vps21Δ ypt7Δ double mutant cells APs are unsealed . The fact that the vps21Δ phenotype masks the ypt7Δ phenotype indicates that Vps21 functions upstream of Ypt7 in the same pathway . We have previously shown that other known components of the endocytic Vps21 GTPase module , the Vps9 GEF , and two of its known effectors , Vps8 , a subunit of the CORVET tethering factor , and the Pep12 SNARE , are also required for autophagy , and APs accumulate upon their depletion [16] . Here , the protease protection assay was used to determine whether , as in vps21Δ , the cargo in APs that accumulate in vps9Δ , vps8Δ and pep12Δ mutant cells is accessible to PK . Both prApe1 and GFP-Atg8 were not protected from degradation in cellular membranes ( P5 ) isolated from vps9Δ and vps9Δ ypt7Δ mutant cells ( Fig 2C and 2D ) , indicating that , like Vps21 , its Vps9 GEF is also involved in AP sealing . In addition , while autophagy cargos prApe1 and GFP-Atg8 were protected in cellular membranes isolated from mutant cells depleted for the Ypt7 effector Vps39 ( 70 and 50% , respectively ) , they were not protected in cellular membranes isolated from mutant cells depleted for the Vps21 effectors , Vps8 and Pep12 ( Fig 2E and 2F ) . Together , these results suggest that components of the Vps21 GTPase module are required for sealing of APs , and show that this step precedes Ypt7-dependent AP fusion with the vacuole . Using three different parameters , our previous studies provided evidence that the APs that accumulate next to the vacuolar membranes in vps21Δ mutant cells are outside the vacuole: 1 ) High magnification fluorescence microscopy; 2 ) EM; and 3 ) time-lapse microscopy which shows that individual APs move throughout the cytosol [16] . While we observed that multiple APs accumulate in clusters adjacent to the vacuolar membrane in vps21Δ mutant cells under autophagy-inducing conditions , Nickerson et al . , proposed that some APs in vps21Δ ypt52Δ mutant cells are inside the vacuoles during normal growth , albeit in a minor fraction of cells ( <10% ) [15] . A similar observation was made in Drosophila cells depleted for Rab5 ( d2 ) , which accumulate Atg8 near the lysosomal membrane , apparently inside the lysosome [29] . Additionally , it has been suggested that Rab5 plays a role in lysosomal function in murine liver cells [30] . Thus , the accumulation of APs in vps21Δ mutant cells could potentially happen adjacent to the vacuolar membrane inside the vacuole [15] . However , we find this unlikely because yeast cells defective in vacuolar proteases , e . g . , Pep4 , accumulate autophagy bodies ( ABs ) uniformly distributed inside their vacuoles [31] . Importantly , the AP localization and accumulation in vps21Δ looked different from the known vacuolar accumulation of ABs in pep4Δ mutant cells by either fluorescence microscopy observation or TEM ( S4A and S4B Fig and [16] ) . Furthermore , the results from the protease protection assay show that autophagy cargos in vps21Δ mutant and vps21Δ pep4Δ double mutant cell lysates were not protected from proteases , while they were protected in pep4Δ mutant cell lysates ( S4C and S4D Fig ) to a level similar to that found in ypt7Δ and vps39Δ mutant cell lysates ( Fig 2 ) . These results imply that the accumulated APs in vps21Δ and vps21Δ pep4Δ mutant cells are unsealed and are outside the vacuole , whereas accumulated ABs in pep4Δ mutant cells are sealed and are inside vacuolar membranes . Taken together , the aforementioned results indicate that the function of Vps21 in autophagy precedes those of Ypt7 and Pep4 , whose depletion results in AP accumulation in the cytoplasm and AB accumulation inside the vacuole , respectively . One hallmark of AP maturation is the dissociation of several Atgs from the AP , e . g . , Atg2 , Atg5 and Atg18 . In contrast , some Atg8 remains attached to APs as they fuse with the vacuole [13] . We wished to determine whether Atgs are still present on unsealed APs that accumulate in cells depleted for Vps21 . The co-localization of Atg2 , Atg5 and Atg18 with the AP marker Atg8 was determined in vps21Δ , ypt7Δ , and vps21Δ ypt7Δ mutant cells under nitrogen starvation . Whereas Atg2 , Atg5 and Atg18 were present only on ~20% of APs that accumulate in cells depleted for Ypt7 , they were present on ~60% of APs and AP clusters that accumulate in vps21Δ and vps21Δ ypt7Δ mutant cells ( Fig 3 ) . Like Atg2 , Atg5 and Atg18 , Atg11 and Atg17 also remained on most APs that accumulate in vps21Δ and vps21Δ ypt7Δ mutant cells ( 60–80% ) , but significantly less were present on APs in ypt7Δ mutant cells ( ~30–40% ) . In addition , Atg11 and Atg17 also remained on the majority of APs in vps9Δ , vps9Δ ypt7Δ , vps8Δ and pep12Δ mutant cells , but not in vps39Δ mutant cells ( Fig 4 ) . Together , these results show that Atg dissociation is defective in mutant cells depleted for Vps21 , its Vps9 GEF or its Vps8 and Pep12 effectors , but not in cells depleted for Ypt7 or its Vps39 effector . Moreover , double mutant analyses support the idea that both Vps9 and Vps21 function upstream of Ypt7 in autophagy . The PI3P phosphatase Ymr1 was previously demonstrated to play a role in Atg dissociation during AP maturation . Specifically , upon addition of rapamycin , ymr1Δ exhibit a defect in general autophagy , accumulation of sealed APs in the cytoplasm , and a failure of Atg dissociation from these APs [18] . We wished to explore the relationship between the roles of Vps21 and Ymr1 in autophagy and to determine whether they function in the same pathway sequentially . This is especially important since the autophagic phenotypes of both vps21Δ and ymr1Δ are partial [16 , 18] . The autophagy defects of the two single-deletion mutant strains and those of the double-deletion mutant strain , were compared upon induction of generic autophagy under nitrogen starvation . Analyses of autophagy cargos processing during nitrogen starvation show that in the single mutant cells , some of the prApe1 and GFP-Atg8 is processed to mApe1 and GFP , respectively . Whereas ymr1Δ mutant cells exhibit a less severe mApe1 processing defect than vps21Δ mutant cells , GFP-Atg8 processing is similar in both mutant cells . Importantly , the autophagy phenotypes of the vps21Δ ymr1Δ double mutant cells do not exceed those of the single deletion strains ( S5 Fig ) . These results are consistent with the idea that Vps21 and Ymr1 function in the same pathway . We next explored AP accumulation under nitrogen starvation in cells depleted for Vps21 , Ymr1 , or both . We have previously shown that APs accumulate in clusters next to the vacuole in vps21Δ mutant cells [16] . AP accumulation was determined here using two approaches , live-cell fluorescence microscopy and electron microscopy ( EM ) . As previously shown [18] , like in ypt7Δ , multiple GFP-Atg8 marked APs are dispersed in the cytoplasm of ymr1Δ mutant cells , and only about 20% of the cells have a GFP-Atg8 cluster . Interestingly , as seen by the FM4-64 staining , unlike in ypt7Δ , vacuoles in ymr1Δ mutant cells are not fragmented . This observation indicates that dispersal of APs in the cytoplasm observed in ymr1Δ and ypt7Δ mutant cells is not caused by vacuole fragmentation . In agreement with the live-cell fluorescence microscopy analysis , AP accumulation in clusters was observed in ~60% of vps21Δ , compared to <5% in ymr1Δ mutant cells using EM . Importantly , AP cluster accumulation in vps21Δ ymr1Δ double mutant cells is similar to that observed in vps21Δ , and not in ymr1Δ , single mutant cells in both assays ( Fig 5A–5C ) , supporting the idea that Vps21 functions upstream of Ymr1 in autophagy . It was previously shown that some prApe1 is protected from proteases in APs isolated from ymr1Δ mutant cells , suggesting that APs that accumulate in these cells are sealed [18] . To further dissect the relationship between depletion of Vps21 and Ymr1 , single and double mutant cells were tested by the protease protection assay using two cargos . As in ypt7Δ , ~50% of prApe1 and GFP-Atg8 was protected from the protease in ymr1Δ mutant cells . In contrast , in vps21Δ ymr1Δ double mutant cells , as in vps21Δ , both autophagy cargos were not protected from the protease ( Fig 5D–5E ) . These results show that Vps21 functions prior to Ymr1 . A defect in Atg dissociation from APs that accumulate in ymr1Δ mutant cells was previously reported [18] . Because we show here that vps21Δ mutant cells are also defective in Atg dissociation from APs , we expected that if Vps21 and Ymr1 function in the same pathway , vps21Δ ymr1Δ double mutant cells will show a phenotype similar to and not exceeding that of the single deletion strains . Indeed , in live-cell fluorescence microscopy analyses , Atg5 , Atg2 , Atg18 and Atg11 were co-localized with APs and AP clusters that accumulate in vps21Δ ymr1Δ mutant cells ( Fig 5F and S6 and S7 Figs ) . These results agree with the idea that Vps21-dependent AP sealing shown above , precedes Atg dissociation during AP maturation , which is defective in cells depleted for Vps21 and/or Ymr1 . PI3P is required for early autophagy [32] , and its removal in later steps was inferred from the role of the PI3P phosphatase Ymr1 after AP accumulation [18] . However , PI3P removal from APs was not shown directly in the later study . Here , we explored PI3P presence on Atg8-marked APs using the PI3P reporter DsRed-FYVE domain [33] and live-cell fluorescence microscopy . This is important since PI3P also decorates endosomes [34] . Co-localization of DsRed-FYVE domain with GFP-Atg8 showed that while PI3P was present only on ~5% of cells with APs that accumulate in ypt7Δ mutant cells , it was present on ~50% of cells with APs or AP clusters that accumulate in vps21Δ , vps21Δ ymr1Δ , and ymr1Δ mutant cells ( Fig 6A and 6B ) . These results show that PI3P removal from APs is defective in cells depleted for Vps21 and/or Ymr1 . In addition , these results agree with the idea that Vps21-dependent AP sealing shown above , also precedes Ymr1-dependent PI3P removal during AP maturation , which is defective in cells depleted for Vps21 and/or Ymr1 . The above mentioned double-mutant analyses suggest that the Vps21-dependent step precedes the Ymr1-dependent step in the autophagy pathway . Therefore , we wished to determine whether Ymr1 localization to APs is dependent on Vps21 . It was previously reported that the majority ( ~75% ) of RFP-Ape1 marked APs co-localize with GFP-tagged Ymr1 during autophagy in wild type cells [18] . We compared this co-localization in wild type and vps21Δ mutant cells expressing Ymr1-yEGFP and mCherry-Ape1 as an AP marker . Whereas Ymr1 co-localized with Ape1 in ~80% of wild type cells , it does so only in ~15% of vps21Δ mutant cells ( Fig 6C and 6D ) . This result suggests that the localization of Ymr1 to APs is dependent on Vps21 and further supports the order of their function , which is based on double mutant analyses . Together , our results show that Vps21 and Ymr1 function sequentially in two separate steps of AP maturation: during or upstream of AP sealing , and PI3P hydrolysis-dependent Atg dissociation from closed APs , respectively . As for the order of function of Ymr1 and Ypt7 , while ymr1Δ mutant cells accumulate dispersed APs like ypt7Δ mutant cells , unlike ypt7Δ mutant cells , they accumulate APs decorated with Atgs and PI3P .
AP maturation , which is required for AP fusion with the vacuole , includes membrane sealing and removal of most Atgs . Results presented here suggest that the Rab5-related Vps21 , in the context of its endocytic GEF-GTPase-effector module , regulates AP closure . Specifically , while cells depleted for members of the Vps21 GTPase module accumulate APs , the cargo enclosed in these APs is not protected from degradation by proteases , implying that these APs are not sealed . AP accumulation next to , but outside of , the vacuole , is supported by microscopy and epistasis analyses showing that Vps21 functions upstream of both Ymr1 and Ypt7 , which accumulate sealed APs dispersed in the cytoplasm . Currently , the evidence that APs accumulating in vps21Δ mutant cells are unsealed is provided by one approach , the protease protection assay . Future higher-resolution morphological analyses should be performed to confirm this idea . Notably , the observation that Atg dissociation , which follows AP closure , is also defective in mutant cells depleted for members of the Vps21 GTPase module , further supports a role of this module in AP maturation . To our knowledge , this is the first report that identifies regulators that function between AP expansion and closure . In addition , using double mutant analyses , we demonstrate that Vps21 functions in autophagy downstream of Ypt1 and upstream of Ypt7 . This establishes a cascade of three successive steps in autophagy regulated by Rab GTPases: Ypt1-mediated AP formation , Vps21-regulated AP closure , and Ypt7-dependent AP fusion with the vacuole ( Fig 6E ) . Interestingly , while Ypt1 regulates AP formation in the context of an autophagy-specific module [8] , the functions of Vps21 and Ypt7 in autophagy are performed together with their cognate GEFs and effectors that also mediate endocytosis [13 , 16] . The possible reason for this discrepancy is that whereas Ypt1 regulates delivery of different cargos to the secretory and the autophagy pathways , Vps21 and Ypt7 regulate preparation of membranes surrounding endosomes or APs for fusion with the same organelle , the lysosome . Because in endocytosis Vps21 and Ypt7 also function sequentially , the endocytic and autophagy pathways seem to converge at the Vps21-mediated step . Previously , the PI3P phosphatase Ymr1 was shown to regulate Atg dissociation from APs [18] . While PI3P has been shown to play a major role early in the autophagy pathway [19 , 32] , it can be inferred from the involvement of Ymr1 that PI3P removal from APs might be required for the completion of this pathway , specifically for Atg dissociation . Because depletion of either Vps21 or Ymr1 does not result in a complete autophagy block [16 , 18] , it was important to determine whether they function sequentially in the same pathway or in parallel pathways . Double mutant and localization analyses shown here establish that Vps21 and Ymr1 function sequentially in AP closure and Atg dissociation , respectively . Moreover , using a PI3P reporter , we show that PI3P is present on APs in cells depleted for Vps21 and/or Ymr1 , but not for Ypt7 . Together , these results demonstrate that Vps21 and Ymr1 regulate the two successive steps of AP maturation: AP closure and PI3P-dependent Atg dissociation , respectively ( Fig 6E ) . What could be the mechanism of Vps21-mediated AP closure ? The fact that Vps21 functions in autophagy and endocytosis in the context of the same module might provide some clues . In endocytosis , Vps21 mediates the formation of late endosomes by the recruitment of the tethering complex CORVET and the Pep12 t-SNARE . Tethering factors and SNAREs mediate membrane fusion . However , it is currently unclear which fusion event is regulated by Vps21 and its effectors in endocytosis . One possibility is that they regulate the homotypic fusion of endosomes to create the large late endosome that contains multiple intra-luminal vesicles ( ILVs ) [10 , 35] . Similarly , it is possible that in autophagy , Vps21 together with its effectors mediate homotypic fusion of APs . However , the size of APs that accumulate in vps21Δ mutant cells , which is within the 400–900 nm range reported for AP size [2] , argues against this possibility . Another possibility is that in endocytosis Vps21 mediates maturation of early endosomes to late endosomes , which are also called multi-vesicular bodies ( MVBs ) . In this process , the early endosomal membrane invaginates to form pouches , and sealing of these “pouches” results in formation of ILVs inside the MVBs [36] . In autophagy , while sealing of the double membrane APs does not require membrane invagination because the whole AP resembles “a single pouch” , the topology of its sealing is similar to that of ILV sealing [13] . Therefore , it is tempting to propose that Vps21 , together with its effectors , CORVET ( e . g . , Vps8 ) and SNAREs ( e . g . , Pep12 ) , mediate membrane scission of the open pouch to create sealed APs [37] . Formally , it is possible that the Vps21 module regulates a yet unknown step that follows AP expansion and precedes AP closure . Regardless , in both endocytosis and autophagy , subsequent progression to the Ypt7-mediated fusion of endosome or sealed AP with the lysosome is expected to follow . In mammalian cells a number of Atgs were implicated in AP closure . First , overexpression of an inactive mutant Atg4B , which interferes with the lipidation of LC3/Atg8 , results in accumulation of incomplete APs [38] . Second , excess of sphingolipids , which disturbs Atg9A trafficking , results in abnormally swollen and unclosed APs [39] . Third , using depletion analyses of LC3 , GABARAP , Atg conjugation systems , and the Atg2-binding protein EPG-6 , a role for these proteins was suggested in AP closure [21 , 40 , 41] . In all these cases , the culprit , the mammalian Atg8 , Atg9 , GABARAP , Atg conjugation systems , and EPG-6 , plays a known role in AP formation or expansion [42] , and , therefore , the effect on AP closure is not specific to this step and can be indirect . Because neither Vps21 nor Ypt7 is required for AP formation and/or expansion , their role in autophagy , e . g . , AP closure and fusion , seems to be specific to a late step of autophagy , and possibly a direct role . Because these players are conserved , we expect that the role of Rab5 in AP closure will pertain to mammalian cells . Currently , while a role for Rab5 in autophagy is accepted [17] , the specific step in which it was implicated varies from AP formation [43] to a late step after that regulated by Rab7 [29] . Future studies should elucidate mechanisms that underlie AP closure and reveal whether , like AP formation and fusion with the lysosome , they are conserved from yeast to humans .
Yeast strains and plasmids used in this study are listed in S1 Table . All yeast and Escherichia coli transformations were as previously described [44] . Construction of strains used for live-cell microscopy of Atg11-3XGFP and mCherry-Atg8 in ypt1-1 related cells: Atg11-3GFP-PG5 plasmid was linearized and integrated in SEY6210 as previously described [45] . This strain was mated with ypt1-1 mutant cells [8] for dissection to get Atg11-3XGFP tagged wild type ( WT ) and ypt1-1 mutant cells . The genotypes of the latter strain was validated by complementation of the ypt1-1 mutant phenotype by YPT1 ( S1A and S1B Fig ) . VPS21 was deleted with the hygromycin resistance cassette to get Atg11-3XGFP tagged vps21Δ and ypt1-1 vps21Δ mutant cells . Wild type and mutant strains expressing endogenously tagged Atg11-3XGFP were transformed with CUP1p-mCherry-Atg8-415 for examining co-localizations under growth and autophagy inducing conditions . Similarly , p1K-GFP-Atg8-406 plasmid was linearized and integrated in WT and ypt1-1 mutant cells as previously described [45] . VPS21 was deleted from the obtained strains using the hygromycin resistance cassette to get GFP-Atg8 tagged vps21Δ and ypt1-1 vps21Δ mutant cells . Construction of the strains used for live-cell microscopy of Atg-GFP dissociation from mCherry-Atg8 marked APs or dsRed-FYVE on GFP-Atg8 marked APs: ORF of specific gene was deleted from AtgX-GFP ( X for 5 , 2 , 18 , 11 , 17 ) or GFP-Atg8 tagged wild type with a drug resistance cassette ( hygromycin or KanMX ) or the LYS2 gene to get different deletion mutants by PCR amplification and recombination . The AtgX-GFP tagged wild type and mutant strains were transformed with CUP1p-mCherry-Atg8-415 for examining co-localizations under starvation conditions . Alternatively , pRS304-mCherry-Atg8 was linearized and integrated in AtgX-GFP ( X for 5 , 2 , 18 , 11 ) tagged strain in SEY6210 background to get AtgX-GFP mCherry-Atg8 strains . Other mutants were obtained with deletion of ORF from AtgX-GFP mCherry-Atg8 strains as above . Finally , GFP-Atg8 tagged cells were examined with the lipophilic dye FM4-64 [45] to observe Atg8 clusters near the vacuolar membranes or with a DsRed-FYVE plasmid to observe FYVE on GFP-Atg8 marked APs . Plasmids expressing mCherry-Ape1 ( pNS1321 ) and Ymr1-yEGFP ( pNS1603 ) under ADH1 promoter were used for co-localization analysis . pNS1603 was constructed as follows: YMR1 ORF without the stop codon was cloned in p415-yEGFP ( pNS1492 ) using SpeI ( vector ) /AvrII ( insert ) and BspEI sites . The antibodies and chemical reagents used in this project have been previously described [16] . For complementation analyses , cells transformed with an empty vector ( pRS425 , 2μ , LEU2 ) or the plasmid for Ypt1 expression were grown in SD-Leu medium overnight and then spotted onto SD-Leu plates in ten-fold serial dilutions , and incubated at indicated temperatures . For live-cell fluorescence microscopy and biochemical analysis , yeast overnight cultures from YPD or selection medium ( when a plasmid was used ) were inoculated at ~0 . 03 OD600 to grow overnight to reach ~0 . 5 OD600 at 26°C in YPD , re-inoculated at ~0 . 05 OD600 to reach ~0 . 5 OD600 . Autophagy was induced by one of two ways: Either 10 nM rapamycin was added to the YPD medium for 4 hours at 26°C [46] , or the cells were washed and starved in SD-N medium ( 0 . 17% yeast nitrogen base without amino acid and ammonium sulfate with 2% glucose ) at 26°C for 2 hours , as previously described [47] . When indicated , FM4-64 was added to a final concentration of 1 . 6 μM to stain the vacuole during the second hour before collecting the cells . A higher percent of vps21Δ mutant cells with GFP-Atg8 clusters were observed when the cells were grown to a lower OD600 and when they were re-inoculated three times before starvation ( S2 Fig ) . Cells expressing fluorescently-tagged proteins from plasmids and/or the chromosome , or stained by FM4-64 , were examined with a Nikon inverted research microscope Eclipse Ti as previously described [45] or with an UltraVIEW spinning-disk confocal scanner unit ( PerkinElmer , Waltham , MA ) with Z-stack of 13 stacks . More than five fields were visualized for each sample . The percentage of co-localized dots ( clusters in vps21Δ mutant cells under autophagy-inducing conditions ) is based on red dots/clusters quantified from 1–6 fields from each experiment . GFP-Atg8 clusters in different mutant cells were quantified as the percentage of GFP-Atg8 dots/clusters per vacuole . Data is presented as the mean ± standard deviation of each variable from three independent experiments . For co-localization analysis of Ymr1-yEGFP and mCherry-Ape1 , WT ( BY4741 , NSY825 ) and vps21Δ ( from the deletion library , NSY1648 ) yeast strains were co-transformed with two plasmids , pNS1603 and pNS1321 . Transformants were grown to early log phase and then shifted to nitrogen-starvation media for 2 h as previously described [16] before being visualized by fluorescent microscopy using deconvolution Axioscope microscope . The protease protection assay combined with immunoblot analysis was carried out as previously described [25] . Briefly , cells expressing GFP-Atg8 were grown to log phase in YPD medium , induced for autophagy in SD-N medium , and then spheroplasted and lysed . Unbroken cells were removed by a centrifugation at 300 × g and the supernatant was subjected to a 5 , 000 x g spin . The 5 , 000 x g-spin pellet , P5 , was re-suspended in buffer and used for proteinase K and Triton X-100 treatments . Following the protease treatment , proteins were precipitated with TCA for immunoblot analysis using anti-GFP or anti-Ape1 antibodies to determine the levels of degraded and non-degraded proteins . Immunoblot assays were conducted as previously described [45] and repeated at least three times . To separate Atg8-PE from Atg8 , lysates of cells expressing HA-tagged Atg8 ( or Atg8ΔR; integrated into the URA3 locus ) were examined on a 13 . 5% SDS-PAGE gel with 6 M urea , and run at a constant current of 25 mA; blots were probed with anti-HA as previously described [48] . Immunoblot bands were quantified using the IMAGEJ software ( National Institutes of Health , USA ) . The percentage of non-degraded GFP-Atg8 was calculated as GFP-Atg8 / ( GFP-Atg8 + GFP ) ×100% . The percentage of non-degraded prApe1 was calculated as prApe1 / ( prApe1 + mApe1 ) ×100% . The percentage of processed GFP was calculated as GFP / ( GFP-Atg8 + GFP ) ×100% . The percentage of mature Ape1 was calculated as mApe1 / ( prApe1 + mApe1 ) ×100% . The protein levels of Atg11-3xGFP and GFP-Atg8 in mutant cells were adjusted to the loading control and compared to the WT levels ( set as 100% ) . Bands from anti-G6PDH antibody were used as a loading control . The data are presented as the mean ± standard deviation of each variable from three independent experiments . The IBM SPSS Statisitcs was applied for significant analysis . For individual samples with ≥ three repeats , data were applied to Analysis Of Variance ( ANOVA ) analysis . If the data passed the Test of Homogeneity of Variances with significance ( p>0 . 05 ) and ANOVA significance ( p<0 . 01 ) , then the data was subjected to Post Hoc Tests with least significant difference ( LSD ) or Duncan analysis to receive the significance of p values . Otherwise , if the data failed the Test of Homogeneity of Variances with significance ( p<0 . 05 ) , then the data was subjected to Post Hoc Tests with Dunnett T3 analysis to receive the p values . P values for relevant comparisons are represented as: n . s . , not significant; * , p <0 . 05; ** , p<0 . 01 . Cells were grown , processed and quantified for transmission electron microscopy as previously described [16] . | In autophagy , a cellular recycling pathway , the double-membrane autophagosome ( AP ) engulfs excess or damaged cargo and delivers it for degradation in the lysosome for the reuse of its building blocks . While plenty of information is currently available regarding AP formation , expansion and fusion , not much is known about the regulation of AP closure , which is required for fusion of APs with the lysosome . Here , we use yeast genetics to characterize a novel mutant phenotype , accumulation of unsealed APs , and identify a role for the Rab5-related Vps21 GTPase in this process . Rab GTPases function in modules that include upstream activators and downstream effectors . We have previously shown that the same Vps21 module that regulates endocytosis also plays a role in autophagy . Using single and double mutant analyses , we find that this module is important for AP closure . Moreover , we delineate three Rab GTPase-regulated steps in the autophagy pathway: AP formation , closure , and fusion , which are regulated by Ypt1 , Vps21 and Ypt7 , respectively . This study provides the first insight into the mechanism of the elusive process of AP closure . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] | [
"cell",
"death",
"autophagic",
"cell",
"death",
"lysosomes",
"vesicles",
"vacuoles",
"enzymes",
"cell",
"processes",
"enzymology",
"light",
"microscopy",
"fungi",
"microscopy",
"cellular",
"structures",
"and",
"organelles",
"endosomes",
"research",
"and",
"analysis",
"methods",
"proteins",
"fluorescence",
"microscopy",
"guanosine",
"triphosphatase",
"yeast",
"biochemistry",
"hydrolases",
"eukaryota",
"cell",
"biology",
"biology",
"and",
"life",
"sciences",
"proteases",
"organisms"
] | 2017 | A Rab5 GTPase module is important for autophagosome closure |
Cadherins are homophilic cell-cell adhesion molecules whose aberrant expression has often been shown to correlate with different stages of tumor progression . In this work , we investigate the interaction of two peptidomimetic ligands with the extracellular portion of human E-cadherin using a combination of NMR and computational techniques . Both ligands have been previously developed as mimics of the tetrapeptide sequence Asp1-Trp2-Val3-Ile4 of the cadherin adhesion arm , and have been shown to inhibit E-cadherin-mediated adhesion in epithelial ovarian cancer cells with millimolar potency . To sample a set of possible interactions of these ligands with the E-cadherin extracellular portion , STD-NMR experiments in the presence of two slightly different constructs , the wild type E-cadherin-EC1-EC2 fragment and the truncated E-cadherin- ( Val3 ) -EC1-EC2 fragment , were carried out at three temperatures . Depending on the protein construct , a different binding epitope of the ligand and also a different temperature effect on STD signals were observed , both suggesting an involvement of the Asp1-Trp2 protein sequence among all the possible binding events . To interpret the experimental results at the atomic level and to probe the role of the cadherin adhesion arm in the dynamic interaction with the peptidomimetic ligand , a computational protocol based on docking calculations and molecular dynamics simulations was applied . In agreement with NMR data , the simulations at different temperatures unveil high variability/dynamism in ligand-cadherin binding , thus explaining the differences in ligand binding epitopes . In particular , the modulation of the signals seems to be dependent on the protein flexibility , especially at the level of the adhesive arm , which appears to participate in the interaction with the ligand . Overall , these results will help the design of novel cadherin inhibitors that might prevent the swap dimer formation by targeting both the Trp2 binding pocket and the adhesive arm residues .
Classical cadherins constitute a subfamily of calcium-dependent cell–cell adhesion proteins that belong to the large and phylogenetically diverse cadherin superfamily . The various members of the classical cadherin subfamily show a tissue-dependent expression profile as well as a high sequence and structure homology . In the different tissues , they are mostly localized at the adherens junctions , where they promote cell–cell adhesion through the homodimeric engagement of ectodomains protruding from neighbouring cells [1 , 2] . This process , which involves cadherin clusterization on the cell membrane , results in the formation of tight cell-cell adhesion interfaces . The extracellular portion of classical cadherins features five tandemly arranged immunoglobulin-like domains ( EC1-EC5 ) , whose relative orientation is controlled and rigidified through the coordination of calcium ions at the interdomain level . The interaction of their cytoplasmic tail with catenins allows a number of cell signalling and trafficking processes , providing also a physical link between cadherins and the actin cytoskeleton machinery [3] . The dynamic adhesive interface of the extracellular portion of E-cadherin and other classical cadherins has been revealed by several crystal structures , which so far have captured only some of the numerous conformational states of the protein [4–8] . In essence , dimerization has been shown to involve mainly the two most membrane-distal domains , EC1 and EC2 . In order to form the so-called ‘strand-swapped dimer’ , two E-cadherin molecules mutually exchange their N-terminal sequence ( the A*-strand or adhesion arm ) , anchoring the aromatic side chain of their Trp2 residue into each other’s binding pocket . Interestingly , to reach the strand swap dimer conformation , which is the reversible endpoint of the dimerization trajectory , two monomeric cadherins must first go through an X-dimer conformation , which lowers the energy of the strand exchange process by firmly placing the two adhesion arms in close physical proximity [9–10] . Beside its broad-ranging effects on physiological tissue organization , classical cadherin dysfunction is often correlated with cancer progression and metastasis [11] . Despite Epithelial ( E ) - to Neural ( N- ) cadherin switching being considered a molecular hallmark of epithelial to mesenchymal transition of cancer cells , carcinomas and distal metastases often retain E-cadherin expression [12] , as observed for instance in late stage tumors in epithelial ovarian cancer ( EOC ) [13–15] . In this context , targeting the E-cadherin adhesive interface with small molecules could have a therapeutic and diagnostic value . Linear and cyclic peptides based on the His79-Ala80-Val81 conserved sequence of the EC1 domain ( not belonging to the swap dimer interface ) have been previously developed to inhibit N-cadherin mediated processes . The most studied cyclic peptide is ADH-1 ( Ac-CHAVC-NH2 ) , which has been shown to be an anti-angiogenic agent able to cause cell apoptosis [16] . Peptidomimetics of ADH-1 have also been identified [17] . Recently [18] , we reported the first library of small peptidomimetic molecules targeting the strand-swapped dimer interfaces of E- and N-cadherin . These compounds are mimics of the tetrapeptide sequence Asp1-Trp2-Val3-Ile4 ( DWVI ) of the adhesion arm , modified by replacing the central dipeptide unit ( WV ) with various scaffolds , all of them bearing a benzyl group that is intended to mimic the indole moiety of Trp2 . The most promising compounds identified by a docking protocol were synthesized and were shown to inhibit cadherin homophilic adhesion in EOC cells at low millimolar concentrations . Of these , compound 1 ( a . k . a . FR159 ) ( Fig 1 ) , was co-crystallized with a mutant of the E-cadherin EC1-EC2 fragment lacking the first two residues of the adhesion arm ( Asp1 and Trp2 ) , while attempts to co-crystallize both compounds 1 and 2 with intact E-cadherin-EC1-EC2 did not yield results [19] . In the X-ray complex structure ( PDB code: 4ZTE ) , compound 1 binds to a dimeric conformation of E-cadherin , the so-called ‘X-dimer’ , which is a key intermediate along the E-cadherin dimerization trajectory leading from the monomeric to the strand swap dimer conformation . Unexpectedly , the ligand was found to occupy a hydrophobic pocket that is formed at the X-dimer interface and not in the Trp2 cavity for which it had been designed . This novel hydrophobic pocket formed by the side chains of residues Ile4 , Pro5 , Ile7 , and Val22 from both cadherin molecules does not overlap with the swap dimer interface . However , in the course of the complex cadherin dimerization mechanism the protein undergoes large conformational changes ( such as for instance the opening of the adhesion arm that leads to strand swap dimer formation ) and goes through different intermediate steps . Hence , it is conceivable that during the process the ligand may bind transiently and via variable moieties also to different surface areas of the protein . Here , to investigate the extent by which variable ligand-cadherin interactions may form over time in solution and to provide a possible dynamic picture of the binding event , we have applied a combination of NMR ( Nuclear Magnetic Resonance ) and computational techniques to the complexes of two peptidomimetic inhibitors from our library ( Fig 1 ) and human E-cadherin-EC1-EC2 . We used ligand-based NMR techniques ( Saturation Transfer Difference , STD , and transferred NOE , tr-NOESY ) [20 , 21] to assess binding occurrence as well as to identify the binding epitope of the ligands and to estimate the dissociation constant of the complexes . The experiments were performed using wild type ( wt ) human E-cadherin-EC1-EC2 at three different temperatures ( 283 , 290 and 298 K ) . Furthermore , we analysed both ligands in the presence of the EC1-EC2 mutant that was used for the X-ray study and that lacks the first two N-terminal residues ( E-cadherin- ( Val3 ) -EC1-EC2 ) . Then , in order to rationalize the atomic details of peptidomimetic-cadherin interaction , the NMR data obtained in the presence of wt E-cadherin-EC1-EC2 were analyzed computationally . Indeed , based on NMR data , docking calculations into the EC1 domain of E-cadherin were carried out first . Then , to take into account the temperature factor and introduce protein flexibility , Molecular Dynamics ( MD ) simulations were also performed . Two temperatures , 300 K and 320 K , were used starting from the different docking poses of 1 and 2 into the E-cadherin model . Of note , ligands binding seems to be highly dependent on the protein flexibility , particularly of the adhesive arm . Overall , the dynamic data described herein will help the design of novel cadherin inhibitors that may bind more efficiently and more selectively into the Trp2 binding pocket .
STD-NMR is a consolidated technique [20 , 21] based on Overhauser effect that is used to study the interactions between small ligands and macromolecules [22–24] . The method relies on the selective irradiation of the protein , which allows magnetization to be transferred to the bound ligand ( which is in great excess in solution compared to the protein concentration ) . The saturated ligand is displaced in solution due to the binding equilibrium and the observation of the ligand signals in the NMR spectrum provides an indication of the interaction . Those ligand protons that are nearest to the protein are more likely to become highly saturated , and therefore show the strongest signal in the mono-dimensional STD spectrum . Owing to the efficiency of the saturation process , the modulation of the ligand signal intensity is used as an epitope-mapping method to describe the target-ligand interactions . In fact , the intensity of the STD signal ( expressed as absolute STD percentage ) reflects the proximity of the ligand to the protein surface . Therefore , the group epitope mapping obtained provides information about the nature of the chemical moieties of the ligand that are crucial for molecular recognition in the binding site . We also used the STD amplification factors ( STD-AF ) to derive the dissociation constant ( KD ) of the ligand-protein complexes [25–27] . Moreover , we performed tr-NOESY experiments in order to determine the preferred bound conformation of the ligands . In solution , the EC1-EC2 domain fragment can adopt several conformations depending on protein and Ca2+ ion concentration . At 1 mM calcium concentration and at less than 40 μM protein concentration , a monomeric species is observed predominantly , while dimeric forms or even oligomers are present at higher protein concentrations in solution ( 600 μM ) [28] . The calcium ions provide a rigidification of the linker region connecting the EC1 and EC2 domains in the monomer , thus making the dimerization surface available . STD–NMR and tr-NOESY spectra were acquired in 20 mM phosphate buffer at pH 7 . 4 ( with 150 mM NaCl and 1 mM CaCl2 ) and 40 μM EC1-EC2 fragments of E-cadherin . In this condition , a monomeric form is predominant in solution [28] . We performed STD-NMR experiments at 283K , 290K and 298K since lowering the temperature helps to shorten the rotational correlation time of the receptor and to increase the effective magnetization transfer . Interestingly , for both compounds we observed different binding epitopes in relation to temperature variations . Furthermore , we also studied the binding modes of the ligands in the presence of the truncated mutant ( E-cadherin- ( Val3 ) -EC1-EC2 ) . The NMR interaction data of peptidomimetics 1 and 2 with the EC1 domain of E-cadherin were further investigated computationally . In the X-ray structure of compound 1 bound to the deleted E-cadherin- ( Val3 ) -EC1-EC2 construct missing the Asp1-Trp2 N-terminal residues , the ligand fits into a novel site at the interface of the X-dimer assembly rather than occupying the Trp2 hydrophobic pocket as initially assumed based on their design strategy . However , according to the NMR results discussed above , the ligands show different binding epitopes depending on the protein construct used , suggesting that the adhesive arm may indeed also be involved in stabilizing the ligand-cadherin binding . To clarify this further , we performed MD simulations starting from the docking poses of 1 and 2 generated into the Trp2 binding pocket of wt E-cadherin model . To take into account temperature modulation , the MD runs were carried out at 300 K and 320 K . For each ligand , a detailed analysis of ligand protonation states and conformational geometries was carried out ( see SI ) in order to select the most relevant conformations and ionization forms for docking calculations .
In this study , we evaluated the binding properties of two peptidomimetic ligands to the extracellular portion of E-cadherin using NMR spectroscopy combined to molecular docking and molecular dynamics calculations . Both ligands have been previously shown to act as mM inhibitors of E-cadherin-mediated cell adhesion [18] and to the best of our knowledge , they are the first peptidomimetics developed from the N-terminal DWVI sequence of the E-cadherin EC1 domain . In these compounds , the central dipeptide unit ( Trp2-Val3 ) of the tetrapeptide motif have been replaced by two different scaffolds bearing an aromatic group that , in our docking model , inserts into the hydrophobic cavity of Trp2 , thus preventing swap dimer formation . In accordance with this hypothesis , a reconstruction of the free energy profile of the conformational transition of the E-cadherin monomer from its closed inactive state ( with the Trp2 indole moiety intramolecularly docked ) to its open form ( indole moiety solvent exposed ) , has shown that in solution the monomer could significantly populate both the open and closed states , which are almost iso-energetic [34] . However , the recent crystallographic structure of 1 in complex with a deleted form of E-cadherin EC1-EC2 domain ( lacking the N-terminal Asp1-Trp2 residues ) showed also a novel possible mechanism of action based on a different adhesive interface [19] . Indeed , in the X-ray structure the peptidomimetic compound binds to the X-dimer conformation of the protein , a crucial kinetic intermediate of the cadherin dimerization pathway , by inserting into a novel hydrophobic cavity that is formed at the interface of the two interacting cadherins . Clearly , owing to the complexity of the E-cadherin homo-dimerization process and to the dynamic behaviour of the target itself , which undergoes major conformational changes as part of its substrate recognition mechanism , further investigations of ligand-cadherin binding are needed . In this work , to sample a set of possible interactions of compound 1 and 2 with E-cadherin , we carried out STD-NMR experiments in the presence of two slightly different E-cadherin species , the wt-E-cadherin-EC1-EC2 and the E-cadherin- ( Val3 ) -EC1-EC2 fragments . Depending on the protein constructs , a different binding epitope of the ligand and also a different temperature effect on STD signals were observed , both suggesting an involvement of the Asp1-Trp2 sequence over time and among the set of possible binding events . Prompted by these considerations , the ligand binding mode proposed by the docking model that places the compound in the Trp2 cavity making interactions with the adhesive arm portion was selected for further investigations . MD simulations were performed starting from the corresponding docking poses of 1 and 2 into a wt-E-cadherin EC1 model . To assess the temperature modulation , MD runs were performed at 300 K and 320 K and the results compared to the STD spectra . The stability of the docking poses and also the interactions with the E-cadherin pocket , including the adhesive arm , were evaluated and compared at the different temperatures . In the input structures , both ligands insert the aromatic group into the Trp2 pocket . The aromatic moiety remains docked into the hydrophobic pocket also during MD runs at 300 K and 320 K , supporting the presence of the aromatic hydrogens STD signal ( detected at all temperatures ) . In general , the starting ligand binding mode is quite conserved during MD run while the protein displays major fluctuations . In particular , for compound 1 protein flexibility increases with temperature , especially for the adhesive arm residues . As a consequence , at 320 K the contacts and the hydrogen bonds of the STD amide protons with the adhesive arm are not present and only the percentage of NHIle contacts with the pocket residues is enhanced compared to 300 K ( Table 1 ) . At 300 K , NH1 resulted to be more engaged by the adhesive arm residues , followed by NH10 and NHIle and this behavior agrees with the STD spectrum at 283 K . At 320 K the population of the ligand NHIle contacts with the protein increased while the other amide protons lost their contacts with the adhesive arm ( too flexible ) and also reduced the interactions with the residues of the binding site . This trend is in good agreement with the STD spectrum at 298 K where the NHIle signal replaces those of NH1 and NH10 . During MD simulations of compound 2 , only the binding mode type A is stable and the ligand remains bound to the receptor . At 300 K , the interaction of NH19 proton with E-cadherin is stronger than NH2 and this behaviour is in agreement with the STD spectra at 283 K and 290 K , showing higher intensity of NH19 signals compared to NH2 . At 320 K , we observed a slight decrease of NH19 contacts ( Asp1 not present ) and , despite losing the interaction with Trp2 , an increase of NH2 contacts population with the pocket residues . This trend is consistent with the intensity variation observed in the corresponding STD spectra . In conclusion , our MD results on ligand-cadherin binding are in agreement with STD spectra . The simulations at different temperatures could also explain the different ligand binding epitope observed in the presence of wt E-cadherin . In particular , the modulation of the signals seems to be dependent on the protein flexibility , especially the adhesive arm , an active part of the binding site that participate in the interaction with the ligand . Based on these results , the design of novel cadherin inhibitors to target more efficiently and selectively the Trp2 binding pocket , will be focused on the stabilization of ligand interactions towards both the hydrophobic cavity and the adhesive arm residues . Moreover , further NMR studies with labelled E-cadherin constructs ( wt vs . deleted forms ) will be carried out to map the protein residues involved in the interaction with the inhibitors and clarify their mechanism of action . | Classical cadherins are the main adhesive proteins at the intercellular junctions and play an essential role in tissue morphogenesis and homeostasis . A large number of studies have shown that cadherin aberrant expression and/or dysregulation often correlate with pathological processes , such as tumor development and progression . Notwithstanding the emerging role played by cadherins in a number of solid tumors , the rational design of small inhibitors targeting these proteins is still in its infancy , likely due to the challenges posed by the development of small drug-like molecules that modulate protein-protein interactions and to the structural complexity of the various cadherin dimerization interfaces that constantly form and disappear as the protein moves along its highly dynamic and reversible homo-dimerization trajectory . In this work , we study the interaction of two small molecules with the extracellular portion of human E-cadherin using a combination of spectroscopic and computational techniques . The availability of molecules interfering in the cadherin homophilic interactions could provide a useful tool for the investigation of cadherin function in tumors , and potentially pave the way to the development of novel alternative diagnostic and therapeutic interventions in cadherin-expressing solid tumors . | [
"Abstract",
"Introduction",
"Results",
"Discussion"
] | [
"protons",
"chemical",
"bonding",
"protein",
"interactions",
"cadherins",
"materials",
"science",
"protein",
"structure",
"hydrogen",
"bonding",
"physical",
"chemistry",
"cell",
"adhesion",
"proteins",
"chemistry",
"salt",
"bridges",
"nucleons",
"molecular",
"biology",
"adhesives",
"physics",
"biochemistry",
"biochemical",
"simulations",
"cell",
"biology",
"nuclear",
"physics",
"electrochemistry",
"biology",
"and",
"life",
"sciences",
"materials",
"physical",
"sciences",
"computational",
"biology",
"macromolecular",
"structure",
"analysis"
] | 2019 | Exploring E-cadherin-peptidomimetics interaction using NMR and computational studies |
Sister centromere fusion is a process unique to meiosis that promotes co-orientation of the sister kinetochores , ensuring they attach to microtubules from the same pole during metaphase I . We have found that the kinetochore protein SPC105R/KNL1 and Protein Phosphatase 1 ( PP1-87B ) regulate sister centromere fusion in Drosophila oocytes . The analysis of these two proteins , however , has shown that two independent mechanisms maintain sister centromere fusion . Maintenance of sister centromere fusion by SPC105R depends on Separase , suggesting cohesin proteins must be maintained at the core centromeres . In contrast , maintenance of sister centromere fusion by PP1-87B does not depend on either Separase or WAPL . Instead , PP1-87B maintains sister centromeres fusion by regulating microtubule dynamics . We demonstrate that this regulation is through antagonizing Polo kinase and BubR1 , two proteins known to promote stability of kinetochore-microtubule ( KT-MT ) attachments , suggesting that PP1-87B maintains sister centromere fusion by inhibiting stable KT-MT attachments . Surprisingly , C ( 3 ) G , the transverse element of the synaptonemal complex ( SC ) , is also required for centromere separation in Pp1-87B RNAi oocytes . This is evidence for a functional role of centromeric SC in the meiotic divisions , that might involve regulating microtubule dynamics . Together , we propose two mechanisms maintain co-orientation in Drosophila oocytes: one involves SPC105R to protect cohesins at sister centromeres and another involves PP1-87B to regulate spindle forces at end-on attachments .
The necessity of sister kinetochores to co-orient toward the same pole for co-segregation at anaphase I differentiates the first meiotic division from the second division . A meiosis-specific mechanism exists that ensures sister chromatid co-segregation by rearranging sister kinetochores , aligning them next to each other and facilitating microtubule attachments to the same pole [1 , 2] . We refer to this process as co-orientation , in contrast to mono-orientation , when homologous kinetochores orient to the same pole . Given the importance of co-orientation in meiosis the mechanism underlying this process is still poorly understood , maybe because many of the essential proteins are not conserved across phyla . Most studies of co-orientation have focused on how fusion of the centromeres and kinetochores is established . In budding yeast , centromere fusion occurs independently of cohesins: Spo13 and the Polo kinase homolog Cdc5 recruit a meiosis-specific protein complex , monopolin ( Csm1 , Lrs4 , Mam1 , CK1 ) to the kinetochore [3–5] . Lrs4 and Csm1 form a V-shaped structure that interacts with the N-terminal domain of Dsn1 in the Mis12 complex to fuse sister kinetochores [6 , 7] . While the monopolin complex is not widely conserved , cohesin-independent mechanisms may exist in other organisms . A bridge between the kinetochore proteins MIS12 and NDC80 is required for co-orientation in maize [8] . In contrast , cohesins are required for co-orientation in several organisms . The meiosis-specific cohesin Rec8 is indispensable for sister centromere fusion in fission yeast [9] and Arabidopsis [10 , 11] . Cohesin is localized to the core-centromere in fission yeast [12] and mice [13] . In Drosophila melanogaster oocytes , we and others have shown that cohesins ( SMC1/SMC3/SOLO/SUNN ) establish cohesion in meiotic S-phase and show an enrichment that colocalizes with centromere protein CID/CENP-A [14–17] . Like fission yeast and mouse , Drosophila may require high concentrations of cohesins to fuse sister centromeres together for co-orientation during meiosis . In mice , a novel kinetochore protein , Meikin , recruits Plk1 to protect Rec8 at centromeres [13] . Although poorly conserved , Meikin is proposed to be a functional homolog of Spo13 in budding yeast and Moa1 in fission yeast . They all contain Polo-box domains that recruit Polo kinase to centromeres [13] . Loss of Polo in both fission yeast ( Plo1 ) and mice results in kinetochore separation [13 , 18] , suggesting a conserved role for Polo in co-orientation . In fission yeast , Moa1-Plo1 phosphorylates Spc7 ( KNL1 ) to recruit Bub1 and Sgo1 for the protection of centromere cohesion in meiosis I [18 , 19] . These results suggest the mechanism for maintaining sister centromere fusion involves kinetochore proteins recruiting proteins that protect cohesion . However , how centromere cohesion is established prior to metaphase I , and how sister centromere fusion is released during meiosis II , still needs to be investigated . We previously found that depletion of the kinetochore protein SPC105R ( KNL1 ) in Drosophila oocytes results in separated centromeres at metaphase I , suggesting a defect in sister centromere fusion [20] . Thus , Drosophila SPC105R and fission yeast Spc7 may have conserved functions in co-orientation [18] . We have identified a second Drosophila protein required for sister-centromere fusion , Protein Phosphatase 1 isoform 87B ( PP1-87B ) . However , sister centromere separation in SPC105R and PP1-87B depleted Drosophila oocytes occurs by different mechanisms , the former is Separase dependent and the latter is Separase independent . Based on these results , we propose a model for the establishment , protection and release of co-orientation . Sister centromere fusion necessary for co-orientation is established through cohesins that are protected by SPC105R . Subsequently , PP1-87B maintains co-orientation in a cohesin-independent manner by antagonizing stable kinetochore-microtubule ( KT-MT ) interactions . The implication is that the release of co-orientation during meiosis II is cohesin-independent and MT dependent . We also found a surprising interaction between PP1-87B and C ( 3 ) G , the transverse element of the synaptonemal complex ( SC ) , in regulating sister centromere separation . Overall , our results suggest a new mechanism where KT-MT interactions and centromeric SC regulate sister kinetochore co-orientation during female meiosis .
Drosophila has three homologs of the alpha type of mammalian Protein Phosphatase1 ( PP1α/γ ) genes , Pp1-87B , Pp1-96A and Pp1-13C [21] . We focused our studies on the Pp1-87B isoform because it is the only essential gene , is highly expressed during oogenesis , and contributes ~80% of PP1 activity during development [21 , 22] . As Pp1-87B mutations are lethal , tissue-specific expression of an shRNA targeting Pp1-87B was used to define its role in oocytes ( see Methods ) [23] . The ubiquitous expression of an shRNA for PP1-87B using tubP-GAL4-LL7 resulted in lethality , suggesting the protein had been depleted . When PP1-87B was depleted in oocytes using mata4-GAL-VP16 ( to be referred to as Pp1-87B RNAi oocytes ) , we observed two phenotypes . The first was disorganization of the metaphase I chromosomes . In wild-type Drosophila oocytes , meiosis arrests at metaphase I with the chromosomes clustered into a single chromatin mass at the center of the spindle ( Fig 1A ) . In 62% of Pp1-87B RNAi oocytes , the single chromosome mass was separated into multiple groups of chromosomes ( Fig 1A and 1B ) . The second phenotype observed in Pp1-87B RNAi oocytes was precocious separation of sister centromeres , as determined by counting the number of centromere protein CENP-C or CID ( CENP-A ) foci ( see Methods ) [24] . In wild-type oocytes , we observed an average of 7 . 3 centromere foci , consistent with the eight expected from four bivalent chromosomes at metaphase I ( Fig 1A and 1C ) . However , in Pp1-87B RNAi oocytes we observed a significantly higher number of foci ( mean = 12 . 7 ) . This suggests a defect in sister centromere fusion results in their premature separation during metaphase I . To determine whether the separated chromosome mass and centromere separation phenotypes in Pp1-87B RNAi oocytes is caused by a general loss of cohesion , we used heterochromatic FISH probes directed to the pericentromeric regions of each autosome to mark the homologs . In wild type , two FISH foci are typically observed per homologous chromosome pair in metaphase I because of cohesion between sister chromatids ( Fig 1D ) . To determine if pericentromeric cohesion in Pp1-87B RNAi oocytes was affected , we analyzed the number of heterochromatin FISH signals from the dodeca satellite , the most punctate and therefore quantifiable heterochromatic FISH probes . In ord mutant oocytes that lack all cohesion , the oocytes had a significantly higher number of dodeca foci ( mean = 4 . 8 ) compared to wild type ( mean = 2 . 7 , Fig 1E ) . In contrast , the average number of dodeca foci in Pp1-87B RNAi oocytes was not significantly higher than wild type ( Fig 1E; mean = 3 . 0 ) , suggesting that pericentromeric cohesion is intact in Pp1-87B RNAi oocytes . Secondly , we used these FISH probes to test if there were loss of arm cohesion , defined as when the homologs separate and are observed as two FISH foci in separate chromosome masses . We observed that while 62% of Pp1-87B RNAi oocytes ( n = 50 ) had a separated chromosome mass , only 8 . 5% of the homologs had separated ( n = 130 ) . These results suggest that arm cohesion is usually retained when PP1-87B is depleted . Hence , the separated chromosome mass phenotype in Pp1-87B RNAi oocytes is due to intact bivalents failing to organize correctly at the center of the spindle . Based on these FISH results , PP1-87B is only required for maintaining sister centromere cohesion but is dispensable for cohesion of the pericentromeric regions and the chromosome arms in oocytes . To refer to this specific type of cohesion , we will use the term sister centromere fusion . We also observed two defects associated with the defect in sister centromere fusion and a lack of co-orientation in Pp1-87B RNAi oocytes . First , the FISH experiments can detect errors in homologs bi-orientation , defined as when pairs of homologous centromeres are separated towards opposite poles ( Fig 1D ) . In Pp1-87B RNAi oocytes , 5 . 3% of the homologs were mono-oriented , defined as when pairs of homologous centromeres are have moved towards the same pole ( n = 130 vs . nwt = 111 , p = 0 . 016 ) . These results support the conclusion that the sister centromere fusion defect in Pp1-87B RNAi oocytes causes problem for homologous chromosomes to bi-orient . Second , when the sister centromeres that precociously separate dring meiosis I in mouse and yeast , chiasmata can still direct bi-orientation of these homologs , suppressing the consequences of co-orientation defects [9 , 13 , 25] . Therefore , we used a crossover defective mutant , mei-P22 [26] , to generate univalents , and knocked down Pp1-87B in these oocytes . If the precocious sister centromere separation causes a co-orientation defect , we would expect the univalents in mei-P22 , Pp1-87B RNAi oocytes can become bi-oriented . Indeed , we observed that 20% of mei-P22 , Pp1-87B RNAi oocytes had sister chromatids bi-oriented ( n = 15 , Fig 1F ) . These results suggest that PP1-87B is required for sister centromere fusion to facilitate co-orientation in metaphase I of oocytes . Both cohesin-dependent and -independent mechanisms of sister centromere fusion have been described . Therefore , we investigated whether loss of sister centromere fusion depends on cohesin release . In addition to PP1-87B , the kinetochore protein SPC105R was also tested because it is the only other protein known to be required for sister centromere fusion in Drosophila oocytes [27] . To investigate if cohesin is involved in sister centromere fusion , we tested if sister centromere separation in Pp1-87B- and Spc105R- RNAi oocytes depends on known cohesin removal mechanisms by depleting two negative-regulators of cohesin , Wings Apart-like ( wapl ) and Separase ( sse ) . If losing a factor required for cohesin removal rescued the sister centromere separation in Pp1-87B or Spc105R RNAi oocytes , it would suggest the Drosophila sister centromere fusion depends on cohesin . Upon co-expression of wapl shRNA with either Pp1-87B or Spc105R shRNA , the centromeres remained separated ( Fig 2A and 2B ) . While WAPL could be required for cohesion release at anaphase I , these results suggest WAPL is not required for the meiosis I sister centromere separation caused by depletion of PP1-87B or SPC105R . However , the centromeres in wapl , Pp1-87B RNAi oocytes became thread-like instead of punctate ( Fig 2A ) , leading to additional centromere foci when quantified ( Fig 2B ) . The thread-like centromere phenotype suggests that chromosome structure is affected in wapl , Pp1-87B RNAi oocytes , consistent with previous studies that concluded WAPL was involved in regulating chromosome structure [28 , 29] . The separated centromere phenotype was rescued in sse , Spc105R RNAi oocytes ( Fig 2A and 2C; mean = 8 . 1 ) , suggesting that centromere separation in Spc105R RNAi oocytes depends on the loss of cohesins . This is a surprising result because it suggests that Separase is active during meiotic metaphase I [30] . If Separase is active , these results could be explained if SPC105R recruits proteins that protect cohesins from Separase . To test the hypothesis that SPC105R protects cohesins from Separase , we examined the localization of MEI-S332/SGO , which is required to maintain cohesion during meiosis in several organisms [31] . Drosophila orthologue MEI-S332 localizes to centromere and peri-centromeric regions in wild-type meiosis I oocytes , as shown by colocalization and substantial non-overlap distribution with the core centromere ( S1 Fig . ) . While present during meiosis I and useful as a marker for cohesion protection , MEI-S332 only shows defects during meiosis II [32 , 33] , possibly due to redundancy with another factor during meiosis I [34 , 35] . Consistent with our hypothesis , MEI-S332 localization was almost abolished in Spc105R RNAi oocytes ( Fig 3A and 3B ) . While we cannot rule out non-cohesive functions for Separase , the most likely interpretation is that SPC105R is required to recruit proteins that protect cohesins from Separase . On the other hand , different from the result of sse , Spc105R RNAi , the separated centromere phenotype was not rescued in sse , Pp1-87B RNAi oocytes ( Fig 2A and 2C; mean = 13 . 4 ) . Consistent with cohesin-independence of these phenotypes , the localization of MEI-S332 in Pp1-87B RNAi oocytes was not reduced , and in fact , the volume was increased relative to wild-type ( Fig 3A and 3B ) . Aurora B is required for MEI-S332 localization [36] , and although the mechanism is not well understood in Drosophila , our results suggest MEI-S332 localization is promoted by Aurora B but constrained by PP1-87B . These results indicate that sister centromere fusion in Drosophila oocytes is regulated through two different mechanisms: the SPC105R pathway that is sensitive to Separase , and the PP1-87B pathway that is Separase independent . Because the Pp1-87B RNAi phenotype was not suppressed by loss of Separase , we investigated cohesin-independent mechanisms for how PP1-87B regulates sister centromere fusion . A critical initial observation was that the spindle volume of Pp1-87B RNAi oocytes was larger than wild type ( Fig 4A ) . In addition , PP1-87B was found to localize to the oocyte meiotic spindle ( S2 Fig . ) . Based on these observations , we tested the hypothesis that PP1-87B regulates microtubules dynamics by co-depleting proteins known to regulate MT dynamics and KT attachments . Aurora B kinase activity is required for spindle assembly in Drosophila oocytes [37] and can be antagonized by PP1 in other systems [38] . Furthermore , they have opposite phenotypes: both the chromosome mass and sister centromeres precociously separate in Pp1-87B RNAi oocytes but remain together in Aurora B-depleted oocytes [37] . Therefore , we tested whether Aurora B is required for both the chromosome mass and centromere separation phenotypes of Pp1-87B RNAi oocytes . Treatment of metaphase I oocytes ( i . e . those that have assembled a spindle ) with the Aurora B inhibitor , Binucleine 2 ( BN2 ) [39] , caused loss of the spindle ( 65% , n = 29 , Fig 4B and 4C ) , consistent with previous findings that Aurora B is required for spindle assembly [37] . Interestingly , Pp1-87B RNAi oocytes showed partial resistance to BN2 treatment; only 9% had lost the spindle and 50% of oocytes had residual MT around the chromosome mass ( n = 22 , Fig 4B and 4C ) . Regardless of these residual MTs , the increased number of centromere foci in Pp1-87B RNAi oocytes ( mean = 13 . 0 ) was rescued by BN2 treatment to a level ( Fig 4D and 4E , mean = 7 . 4 ) similar to wild-type controls ( Fig 4D and 4E , mean = 7 . 7 ) . Similarly , the increased frequency of chromosome mass separation in Pp1-87B RNAi oocytes was rescued by BN2 treatment ( Fig 4F and 4G ) . In contrast , centromere separation was not rescued by BN2 treatment of Spc105R RNAi oocytes ( Fig 4B and 4C , mean = 11 . 3 ) . These results are concordant with the effects of sse RNAi on the Spc105R and Pp1-87B RNAi phenotypes and support the conclusion that the maintenance of centromere fusion may occur by at least 2 mechanisms . Suppression of Pp1-87B RNAi oocyte phenotypes by BN2 treatment could have been due to loss of Aurora B activity , or loss of the spindle microtubules . To distinguish between these two possibilities , we treated Pp1-87B RNAi oocytes with Paclitaxel to stabilize the spindle prior to BN2 treatment of the oocytes . Although these oocytes successfully formed spindles , 18% showed chromosome mass separation , a significant decrease compared to the Paclitaxel and solvent-treated RNAi control oocytes and similar to the results from BN2 treatment of Pp1-87B RNAi oocytes ( Fig 4F and 4G ) . This rescue of chromosome mass separation demonstrates that PP1-87B antagonizes Aurora B in regulating chromosome organization . On the contrary , the sister centromeres remained separated in these oocytes ( Fig 4D and 4E , mean = 11 . 1 ) , suggesting that stabilizing microtubule dynamics in Pp1-87B RNAi oocytes can override any effect of inhibiting Aurora B on sister centromere fusion . Based on these observations , we propose that PP1-87B regulates sister centromere separation by regulating microtubules dynamics . However , we cannot rule out the possibility that Aurora B is also required for centromere separation independently of the microtubules . The meiotic spindle consists of overlapping microtubules , only a portion of which make contact with the kinetochores . To understand which set of microtubules affect PP1-dependent centromere separation and chromosome mass disorganization , we used knockdowns of kinetochore proteins to specifically abrogate one class of microtubule contracts with the chromosomes . In Drosophila oocytes , SPC105R is required for lateral attachments and the localization of NDC80 whereas NDC80 is required for end-on attachments [20] . Thus , we co-depleted PP1-87B and SPC105R ( no MT attachments ) or NDC80 ( lateral MT attachments only ) and examined the chromosomes and centromeres . We found that loss of SPC105R , but not NDC80 , suppressed the separated chromosome mass phenotype of Pp1-87B RNAi oocytes ( Fig 5A and 5C ) , suggesting that the separated chromosome mass phenotype in Pp1-87B RNAi oocytes depends on lateral KT-MT interactions . The sister centromeres are already separated in Spc105R RNAi oocytes , and co-depletion of both Pp1-87B and Spc105R did not enhance the effects of either single knockdowns ( Fig 5A and 5B ) . In contrast , the centromere separation phenotype was rescued in Ndc80 , Pp1-87B double RNAi oocytes ( mean = 9 . 0 , Fig 5A and 5B ) but not chromosome mass disorganization . We conclude that PP1-87B affects chromosome mass organization through regulating lateral KT-MT attachments and sister centromere fusion through end-on attachments . To identify proteins that function with PP1-87B in regulating end-on KT-MT attachments , we depleted proteins with meiotic functions that are involved in regulating microtubule attachments . Polo kinase localizes to centromeres in Drosophila metaphase I oocytes [40] ( S3 Fig . ) , and in other organisms has been reported to stabilize KT-MT attachments [41–44] . Unlike Polo in mice [13] , Drosophila polo RNAi oocytes do not show precocious sister centromere separation at metaphase I [45] . We depleted polo with RNAi in either Spc105R or Pp1-87B RNAi oocytes . Interestingly , we found that centromere separation in both mutant oocytes were rescued by polo RNAi ( Fig 6A and 6B , mean = 6 . 6 and mean = 6 . 9 ) . These results indicate that Polo negatively regulates both the separase-dependent ( through SPC105R ) and the microtubule attachment dependent pathways ( through PP1-87B ) for sister centromere fusion in Drosophila . Two proteins , BubR1 and MPS1 , function along with Polo to regulate KT-MT attachments in several organisms [41 , 42 , 46 , 47] . We predicted that depletion of either one could have a similar effect on the Pp1-87B oocyte phenotype as polo RNAi . Centromere separation in Pp1-87B RNAi oocytes was suppressed by simultaneous knockdown of BubR1 ( Fig 6A and 6C; mean = 8 . 5 ) but not mps1 ( Fig 6A and 6C; mean = 10 . 7 ) . A caveat to this negative result is that , based on the non-disjunction rate , MPS1 is only partially depleted in these females ( NDJ = 11% , n = 961 , compared to a strong mps1 loss of function mutant , NDJ = 20 . 2% , n = 231 [48] ) . Regardless , these results suggest that PP1-87B promotes sister centromere fusion by antagonizing the activities of Polo and BubR1 . In contrast , the frequency of oocytes with a separated chromosome mass phenotype remained similar to Pp1-87B RNAi oocytes when PP1-87B were co-depleted with BubR1 ( Fig 6D ) , consistent with the results with NDC80 . This result confirms that the separated chromosome mass phenotype in Pp1-87B RNAi oocytes depends on lateral KT-MT interactions . We propose that PP1-87B destabilizes end-on microtubule attachments by antagonizing Polo and BubR1 activities . In support of this conclusion , the increased spindle volume observed in of Pp1-87B RNAi oocytes was suppressed by co-depletion of polo or BubR1 ( Fig 6E ) . In summary , several experiments , including drug treatment ( Paclitaxel+BN2 ) , depletion of genes that affect KT-MT attachments , and measurements of spindle volume , support the conclusion that PP1-87B regulates KT-MT attachments , and these activities then affect sister-centromere separation and chromosome mass organization . As described above , simultaneous loss of co-orientation and chiasmata can result in bi-orientation of univalent at meiosis I . We observed this phenomenon with simultaneous depletion of PP1-87B and mei-P22 . The same experiment was done with c ( 3 ) G , which encodes a transverse element of the synaptonemal complex ( SC ) [49] , because it is also required for crossing over [50] . Compared to mei-P22 , however , we got surprisingly different results . First , c ( 3 ) G mutant females that were depleted of Pp1-87B failed to produce mature oocytes . We currently do not know why loss of c ( 3 ) G and prophase depletion of PP1-87B causes a failure in oocyte development , but it suggests C ( 3 ) G has a function in mid-oogenesis after its role in crossing over . To examine the interaction between C ( 3 ) G and PP1-87B , c ( 3 ) G RNAi was used . To test the efficiency of the c ( 3 ) G RNAi , nanos-VP16-GAL4 was used to express the shRNA during early prophase , the frequency of X chromosome non-disjunction ( NDJ ) was similar to that observed in c ( 3 ) G null alleles ( 31% , n = 1647 ) [50] . In addition , C ( 3 ) G localization was absent in the germarium ( S4 Fig . ) . These results suggest that this shRNA knockdown recapitulates the null mutant phenotype . For the double depletion we used mata4-VP16-GAL that induced shRNA expression later in oogenesis than nanos-VP16-GAL4 . This was necessary because early expression of Pp1-87B shRNA results in a failure to produce oocytes . When using mata4-VP16-GAL to express shRNA , C ( 3 ) G was present in pachytene , crossing over was not affected ( NDJ = 0% , n = 427 ) , but C ( 3 ) G was missing from mid-late prophase ( Fig 7A , S4 Fig . ) . These results indicate C ( 3 ) G is dynamic throughout prophase , and allows us to test if there is a late prophase-metaphase interaction between C ( 3 ) G and PP1-87B . Interestingly , RNAi of c ( 3 ) G rescued the sister centromere separation phenotype in Pp1-87B , but not Spc105R RNAi oocytes ( Fig 7A and 7B ) . These results suggest that PP1-87B antagonizes centromeric C ( 3 ) G , after most of the SC has been disassembled , to maintain sister centromere fusion at metaphase I . As with other proteins that regulate end-on attachments , the Pp1-87B RNAi increased spindle volume phenotype was rescued to wild type levels by co-depletion of c ( 3 ) G ( Fig 7C ) . It is noteworthy that C ( 3 ) G is enriched at the centromere regions [51 , 52] in pachytene , although its function there is not known . In this location , and because C ( 3 ) G has a Polo-binding box , it is possible that C ( 3 ) G recruits Polo to the centromere region to regulate microtubule dynamics . However , when examining the localization of Polo in c ( 3 ) G RNAi oocytes , we did not observe any changes in protein localization compared to wild-type ( S4 Fig . ) . Whether C ( 3 ) G plays a role in regulating microtubule dynamics through Polo or other independent function to regulate sister centromere fusion needs further investigation .
Assembly of meiosis-specific cohesins at the centromeres probably establishes sister centromere fusion [2] . Indeed , the meiosis-specific cohesin complex SMC1/SMC3/SOLO/SUNN is enriched at Drosophila meiotic centromeres and could to have this function [14–17] . Guo et al found that separase is required for progression through both meiotic divisions in oocytes [53] . We found that depleting Separase in metaphase I Drosophila oocytes rescued the precocious centromere separation phenotype caused by loss of SPC105R . Although we found no role for WAPL in centromere fusion , we did not rule out other functions in meiosis , especially in anaphase I given that Guo et al . found that oocytes depleted for Separase were delayed in anaphase I but eventually progressed to meiosis II . SPC105R may protect centromere cohesion by recruiting cohesin protection proteins such as MEI-S332/SGO that subsequently recruit PP2A . The fact that mei-S332 mutants do not display defects in meiosis I [32 , 33] could be due to redundancy with another Drosophila PP2A recruiter , Dalmatian [34 , 35] . The previous finding that Drosophila Spc105R mutants enhance defects in Separase function suggest SPC105R may have a cohesion protection function in other cell types [54] . However , Separase activation usually coincides with the entry into anaphase when the APC degrades an inhibitor of Separase , Securin [30] . One explanation is that Separase has a novel cohesin-independent function in regulating co-orientation through SPC105R , such as structural or regulatory function within the kinetochore or spindle [55] , or that loss of SPC105R activates Separase . We favor , however , the explanation that Separase is active prior to anaphase I and cohesion is maintained only by PP2A activity in metaphase I arrested oocytes . This model can explain why knockout of SPC105R in male meiosis does not show a loss of centromere fusion [56] . In male meiosis where there is no cell cycle arrest , Separase may not be active until anaphase , which would make a protective role for SPC105R difficult to observe . Aurora B inhibitor BN2 was applied to mature Pp1-87B RNAi oocytes , which were in prometaphase I or metaphase I and therefore , after the spindle had formed and the sister centromeres had separated . Because this treatment caused the sister centromeres to come back together , sister centromere separation in Pp1-87B RNAi oocytes appears to be reversible . In contrast , treatment of mature oocytes with BN2 did not reverse centromere separation in Spc105R RNAi . This reversible phenotype of Pp1-87B RNAi oocytes is consistent with a mechanism that involves the reorganization of centromere and kinetochore geometry , and the nonreversible phenotype of Spc105R RNAi with a mechanism that involves the degradation of cohesins . Furthermore , the results from destabilizing microtubule attachments with BN2 treatment suggested that centromere separation in PP1-87B-depleted oocytes depends on KT-MT interactions . In support of this conclusion , we found that PP1-87B affects several spindle-based parameters: it localizes to the meiotic spindle , its knockdown caused an increase in spindle volume , and centromere separation in PP1-87B-deplated oocytes depended on NDC80 , Polo and BubR1 . These results suggest that stable end-on attachments are required for release of sister centromere fusion . Similar conclusions have been made in Drosophila male meiosis . Sister centromere separation in meiosis II does not depend on Separase [57] but does depend on KT-MT interactions [56 , 58] . These findings are not limited to Drosophila . Classic micro-manipulation experiments in grasshopper cells demonstrated that the switch in meiosis II to separated sister kinetochores requires attachment to the spindle [59] . Based on all these results , we propose that sister centromeres normally separate early in meiosis II by a process that is Separase-independent but microtubule-dependent ( Fig 7D ) . Interestingly , univalents in meiosis I can biorient in co-orientation-defective mutants that lack crossovers [13] . We observed a similar phenomenon in Pp1-87B; mei-P22 meiosis I oocytes . However , the frequency of univalent bi-orientation was low , raising the question of how meiosis II univalents preferentially achieve bi-orientation . The low frequency of univalent bi-orientation in meiosis I could be due to differences in how each division begins . Meiosis I begins with the centromeres clustered in a chromocenter and rapidly develops a robust central spindle , both of which may bias the sister centromeres to make attachments to the same pole , even in a PP1-87B knockdown oocyte ( S5 Fig . ) . The mechanism regulated by PP1-87B that regulates KT-MT interactions and maintains sister centromere fusion is not known and may be a function utilized in mitotic cells . For example , PP1 has a role in regulating microtubule dynamic in Xenopus extracts [60] . In HeLa cells depleted of SDS22 , a regulatory subunit of PP1 , sister kinetochore distances increase [61] , similar to the defect we described here . In budding yeast , suppressing premature formation of stable kinetochore-microtubule attachments is necessary for co-orientation [62] . The mechanism may be related to the role of PP1 in negatively regulating condensin functions that affect chromosome structure [63 , 64] . A negative effect on condensin activity , which is known to shape mitotic chromosomes [65 , 66] , could also explain the chromosome mass separation phenotype of Pp1-87B RNAi oocytes . Our observations are strikingly similar to the phenomenon of cohesin fatigue , where sister chromatids separate in metaphase arrested mitotic cells . Identical to the effect of PP1-87B on centromere separation , cohesion fatigue occurs in a Separase-independent but microtubule-dependent manner [67 , 68] , however , the mechanism is unknown [69] . Oocytes with a prolonged arrest points , such as metaphase I in Drosophila , might prevent cohesion fatigue by concentrating meiotic cohesins at the centromeres and destabilizing KT attachments to reduce MT forces . In Drosophila oocytes , the microtubule catastrophe protein Sentin destabilizes end-on KT-MT attachments after the spindle is well established [27] . In fact , active destabilization of kinetochore attachments may be a common feature of oocyte meiosis . Mammalian oocytes also have an extended period of dynamic KT-MT interactions [70] , lasting 6–8 hours in mice and up to 16 hours in human [71 , 72] . All of these results are in line with our conclusion that oocytes require PP1-87B to prevent premature stable KT-MT attachments and avoiding cohesion fatigue . Depletion of C ( 3 ) G suppresses the Pp1-87B centromere fusion defect . This result suggests that centromeric SC has a role in negatively regulating sister centromere co-orientation . While the bulk of SC disassembles in late prophase [73 , 74] , centromeric SC proteins persist beyond pachytene in Drosophila and until at least metaphase I in budding yeast and mouse [51 , 74–77] . It has also been shown that SC proteins interact with the NDC80 complex in two yeast two hybrid experiments [78 , 79] . These studies have concluded that centromeric SC is required for bi-orientation of homologs and monopolar attachment . Because both Polo Kinase and C ( 3 ) G negatively regulate co-orientation , we hypothesize that C ( 3 ) G could be required for Polo Kinase activity , but not localization , at the centromere . Thus , centromeric SC components might be an important mediator of co-orientation . Co-orientation in yeast and mice depends on Polo kinase , which is recruited by Spo13 , Moa1 or Meikin [80] . This is opposite of the known mitotic role of Polo in phosphorylating cohesin subunits and facilitating their removal from binding sister chromatids [31 , 81 , 82] . In yeast meiosis , however , the phosphorylation of cohesin subunits may depend on two different kinases , Casein kinase I and CDC7 [83–85] . Which kinase ( s ) are required in animals to phosphorylate meiotic cohesins for their removal remains unknown . We have shown that Polo is required for loss of centromeric cohesion , which to our knowledge is the first evidence of its kind in animal meiotic cells . Unlike mice and yeast , depletion of Polo kinase from Drosophila metaphase I oocytes does not cause sister centromere separation [45] . One reason for this difference in Polo function could be that it is required at multiple stages of meiosis and its phenotype may depend on when it is absent . Loss of Polo or BubR1 during early Drosophila prophase ( pachytene ) oocytes leads to loss of SC and cohesion defects [86 , 87] . Our experiments depleted Polo after cohesion was established . Alternatively , the function of Polo in co-orientation may not be conserved . Importantly , two features of centromere fusion and co-orientation that are conserved are maintenance depending on SPC105R and separation depending on stabilization of KT-MT attachments . Like SPC105R in Drosophila , budding yeast KNL1 is required for meiotic sister centromere fusion and co-orientation and is a target of Polo [18] . The differences between Drosophila and mouse or yeast can be explained if SPC105R does not require Polo in order to protect cohesion at the centromeres for co-orientation . While all previous studies of co-orientation have focused on the establishment of centromere fusion , our results identified several key regulators and provide insights into how sister centromere fusion is maintained in meiosis I and released for meiosis II . In contrast to release of cohesion in most regions of the chromosomes , we propose a Separase-independent mechanism that requires stable kinetochore-microtubule attachments promotes centromere separation early in meiosis II . While it is well known that regulating microtubule attachments is important for bi-orientation , our results are an example of another reason why KT-MT attachments must be properly regulated , to safely navigate the transitions through the two divisions of meiosis .
Drosophila were crossed and maintained on standard media at 25°C . Fly stocks were obtained from the Bloomington Stock Center or the Transgenic RNAi Project at Harvard Medical School [TRiP , Boston , MA , USA , flyrnai . org , 23] . Information on genetic loci can be obtained from FlyBase [flybase . org , 88] . Most Drosophila lines expressing a short hairpin RNA were designed and made by the Transgenic RNAi Project , Harvard ( TRiP ) ( Table 1 ) . To deplete target mRNA , a cross was performed to generate females carrying both the UAS:shRNA and a GAL4-VP16 transgene . The shRNA can be induced ubiquitous expression by crossing to tubP-GAL4-LL7 and testing lethality [89] , or mata4-GAL-VP16 and osk-GAL4-VP16 for oocyte-specific expression [90] . In this paper , mata4-GAL-VP16 was primarily used for inducing expression of the UAS:shRNA after early pachytene but throughout most stages of oocyte development in the Drosophila ovary . This allows for 3–5 days of continuous expression to knockdown the mRNA levels . In some cases , we used the oskar -GAL4-VP16 transgene [91 , 92] , which causes a similar knockdown and phenotype in PP1-87B as mata4-GAL-VP16 . Double RNAi crosses were set up based on the available RNAi lines ( Table 2 ) . For measuring the mRNA knockdown level , total RNA was extracted from late-stage oocytes using TRIzol Reagent ( Life Technologies ) and reverse transcribed into cDNA using the High Capacity cDNA Reverse Transcription Kit ( Applied Biosystems ) . The qPCR was performed on a StepOnePlus ( Life Technologies ) real-time PCR system using TaqMan Gene Expression Assays ( Life Technologies ) , Dm02152292_g1 for Pp1-87B and Dm02134593_g1 for the control RpII140 . Oocyte-specific shRNA expression of HMS00409 using mata4-GAL-VP16 resulted in sterility and knockdown of the oocyte mRNA to 35% as measured by RT-qPCR; the same phenotype has been seen when using osk-GAL4-VP16 , where the mRNA knockdown is also to 35% . For SPC105R , expressing shRNA GL00392 using osk-GAL-VP16 knocked down the mRNA to 10% . To generate a wapl shRNA line , we followed the protocol in Harvard TRiP center ( http://fgr . hms . harvard . edu/trip-plasmid-vector-sets ) and targeted wapl sequence ( 5’-gaggaggaggatcaacagcaa -3’ ) for mRNA knockdown . This 21-nucleotide sequence was cloned into pVALIUM22 and the whole construct was injected into Drosophila embryos ( y sc v; attP40 ) . The mRNA is knocked down to 4% when using mata4-GAL-VP16 to express the shRNA in oocytes . Mature ( stage 12–14 ) oocytes were collected from 100 to 200 , 3-4-day old yeast-fed non-virgin females . The procedure is described as in [93] . Oocytes were stained for DNA with Hoechst 33342 ( 10 μg/ml ) and for MTs with mouse anti-α tubulin monoclonal antibody DM1A ( 1:50 ) , directly conjugated to FITC ( Sigma , St . Louis ) . Additional primary antibodies used were rat anti-Subito antibody [40] , rat anti-INCENP [94] , guinea pig anti-MEI-S332 [95] , rabbit anti-CENP-C [96] , rabbit anti-Deterin [97] , rabbit anti-SPC105R [98] , mouse anti-Polo [99] and rabbit anti-CID ( Active Motif ) . These primary antibodies were combined with either a Cy3 , Alex 594 or Cy5 secondary antibody pre-absorbed against a range of mammalian serum proteins ( Jackson Immunoresearch , West Grove , PA ) . FISH probes corresponding to the X359 repeat labeled with Alexa 594 , AACAC repeat labeled with Cy3 and the dodeca repeat labeled with Cy5 were obtained from IDT . Oocytes were mounted in SlowFade Gold ( Invitrogen ) . Images were collected on a Leica TCS SP8 confocal microscope with a 63x , NA 1 . 4 lens . Images are shown as maximum projections of complete image stacks followed by merging of individual channels and cropping in Adobe Photoshop ( PS6 ) . All the CENP-C foci , CID foci , chromosome mass volume , spindle volume and MEI-S332 volume were measured using Imaris image analysis software ( Bitplane ) . For determining centromere foci , an automated spots detection function in Imaris was used . A spot whose XY diameter is 0 . 20 μm , Z diameter is 1 . 00 μm and is in touch with DNA will be counted as a centromere . For the volume measurement , images were resampled first to become isovoxel data . Then the surface detection function was used for defining different objects and measuring their volume . Spindle volume is measured and normalized by the chromosome mass volume to compensate for effects of chromatin volume on microtubule recruitment , although chromosome mass volume in each genotype did not differ significantly . To inhibit Aurora B , oocytes were incubated with either 0 . 1% DMSO or 50 μM BN2 in 0 . 1% DMSO for 60 minutes prior to fixation in Robb’s media . To stabilize MTs , oocytes were incubated with either 0 . 1% DMSO or 10 μM Paclitaxel ( Sigma ) in 0 . 1% DMSO for 10 minutes , followed by 50 μM BN2 plus 10 μM Paclitaxel in 0 . 1% DMSO for 60 minutes . Statistical tests were performed using GraphPad Prism software . All the numbers of the centromere foci or spindle/chromosome mass volume or MEI-S332 volume were pooled together and ran one-way ANOVA followed by post hoc pairwise Tukey’s multiple comparison test . Details of statistical evaluations and the numbers of samples are provided in the figure legends . | Meiosis involves two cell divisions . In the first division , pairs of homologous chromosomes segregate , in the second division , the sister chromatids segregate . These patterns of division are mediated by regulating microtubule attachments to the kinetochores and stepwise release of cohesion between the sister chromatids . During meiosis I , cohesion fusing sister centromeres must be intact so they attach to microtubules from the same pole . At the same time , arm cohesion must be released for anaphase I . Upon entry into meiosis II , the sister centromeres must separate to allow attachment to opposite poles , while cohesion surrounding the centromeres must remain intact until anaphase II . How these different populations of cohesion are regulated is not understood . We identified two genes required for maintaining sister centromere cohesion , and surprisingly found they define two distinct mechanisms . The first is a kinetochore protein that maintains sister centromere fusion by recruiting proteins that protect cohesion . The second is a phosphatase that antagonizes proteins that stabilize microtubule attachments . We propose that entry into meiosis II coincides with stabilization of microtubule attachments , resulting in the separation of sister centromeres without disrupting cohesion in other regions , facilitating attachment of sister chromatids to opposite poles . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] | [
"invertebrates",
"meiosis",
"rna",
"interference",
"microtubules",
"chromosome",
"structure",
"and",
"function",
"centromeres",
"metaphase",
"cell",
"cycle",
"and",
"cell",
"division",
"atmospheric",
"science",
"cell",
"processes",
"animals",
"germ",
"cells",
"animal",
"models",
"oocytes",
"drosophila",
"melanogaster",
"model",
"organisms",
"experimental",
"organism",
"systems",
"epigenetics",
"cellular",
"structures",
"and",
"organelles",
"drosophila",
"cytoskeleton",
"research",
"and",
"analysis",
"methods",
"chromosome",
"biology",
"animal",
"cells",
"genetic",
"interference",
"animal",
"studies",
"gene",
"expression",
"atmospheric",
"phenomena",
"aurora",
"insects",
"arthropoda",
"biochemistry",
"rna",
"eukaryota",
"cell",
"biology",
"ova",
"nucleic",
"acids",
"earth",
"sciences",
"genetics",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"organisms",
"chromosomes"
] | 2019 | Sister centromere fusion during meiosis I depends on maintaining cohesins and destabilizing microtubule attachments |
Ehrlichia chaffeensis is a tick transmitted pathogen responsible for the disease human monocytic ehrlichiosis . Research to elucidate gene function in rickettsial pathogens is limited by the lack of genetic manipulation methods . Mutational analysis was performed , targeting to specific and random insertion sites within the bacterium's genome . Targeted mutagenesis at six genomic locations by homologous recombination and mobile group II intron-based methods led to the consistent identification of mutants in two genes and in one intergenic site; the mutants persisted in culture for 8 days . Three independent experiments using Himar1 transposon mutagenesis of E . chaffeensis resulted in the identification of multiple mutants; these mutants grew continuously in macrophage and tick cell lines . Nine mutations were confirmed by sequence analysis . Six insertions were located within non-coding regions and three were present in the coding regions of three transcriptionally active genes . The intragenic mutations prevented transcription of all three genes . Transposon mutants containing a pool of five different insertions were assessed for their ability to infect deer and subsequent acquisition by Amblyomma americanum ticks , the natural reservoir and vector , respectively . Three of the five mutants with insertions into non-coding regions grew well in deer . Transposition into a differentially expressed hypothetical gene , Ech_0379 , and at 18 nucleotides downstream to Ech_0230 gene coding sequence resulted in the inhibition of growth in deer , which is further evidenced by their failed acquisition by ticks . Similarly , a mutation into the coding region of ECH_0660 gene inhibited the in vivo growth in deer . This is the first study evaluating targeted and random mutagenesis in E . chaffeensis , and the first to report the generation of stable mutants in this obligate intracellular bacterium . We further demonstrate that in vitro mutagenesis coupled with in vivo infection assessment is a successful strategy in identifying genomic regions required for the pathogen's in vivo growth .
Ehrlichia chaffeensis , an alpha-proteobacterium , is an intracellular pathogen that is transmitted through an infected Lone Star tick , Amblyomma americanum , to humans and several other vertebrate hosts [1]–[8] . The pathogen is responsible for causing human monocytic ehrlichiosis ( HME ) [6] , [7] , [9] , [10] . The disease is characterized by an acute onset of febrile illness that can progress to a fatal outcome , particularly in immune compromised individuals [2] , [11] . Clinical symptoms of the flu like illness include malaise , nausea , headache , myalgia and persistent fever . Leukopenia , thrombocytopenia , and elevated liver transaminases are common laboratory findings [9] , [10] , [12] . E . chaffeensis and related pathogens have evolved unique strategies to establish infections in both ticks and mammals in order to successfully complete their transmission cycle [13] , [14] . Persistent infection throughout the developmental stages of ticks is important , as the organism cannot be transmitted transovarially to larval offspring . Our recent molecular and proteomic studies have revealed global differences in the expressed proteins of E . chaffeensis within different host cell environments [15]–[18] . The pathogen's differential gene expression in response to distinct cellular environments is a major contributor for its dual host adaptation and persistence [19] . Targeted or random mutagenesis is routinely employed to study gene function in bacteria that can be grown axenically . Transformation of obligate intracellular pathogens , such as E . chaffeensis , remains a challenge , particularly considering its limited viability in the extra-cellular environment [20] . Transformation experiments , therefore , must be conducted during a small window of time when the bacteria must remain competent to infect host cells . Secondly , as Ehrlichia species do not harbor plasmids , the introduced foreign DNA in the form of a plasmid or linear fragments must remain intact during the transformation and infection stages so that the encoded genes can be expressed using the bacterium's RNA polymerase complex . Finally , the choice of an antibiotic resistance cassette to be introduced into the organisms should not target antibiotics useful in treating a patient . In this study , we considered all these aspects in creating mutational methods in E . chaffeensis . Homologous recombination is useful for targeting mutations to a specific genomic site , and has been successfully employed in making gene disruption mutations in several intracellular bacteria [21]–[24] . Similarly , a modified mobile group II intron ( TargeTron ) -based targeted mutagenesis has been proven valuable for efficient gene targeting in several Gram-negative and Gram-positive bacteria [25]–[31] . Recent studies have shown that the Himar1 transposase system is a valuable tool for transposon-mutagenesis of various bacterial organisms , including alpha-proteobacteria closely related to E . chaffeensis , such as Anaplasma phagocytophilum [32] , [33] . In this study , we evaluated methods for creating targeted mutations in E . chaffeensis by employing both homologous recombination and TargeTron methods , and random mariner mutagenesis using the Himar1 transposase system . Six different genomic locations were assessed by targeted mutagenesis which led to the consistent identification of mutants at three genomic sites . We generated nine random transposon-mediated mutations in the E . chaffeensis genome , three of which disrupted the coding regions of different transcriptionally active hypothetical protein genes , and six in intergenic sites . Four of the insertions also caused loss of gene expression . We present the first evidence that insertion mutations at three sites within the E . chaffeensis genome abolished the growth of the organism in its natural host .
We evaluated the ability of spectinomycin , rifampin , chloramphenicol , gentamicin , and kanamycin and ampicillin to inhibit the growth of E . chaffeensis in the canine macrophage cell line , DH82 . With the exception of ampicillin , all of the antibiotics inhibited the growth of the organism . The minimum concentrations of antibiotics required for 100% growth inhibition by day 7 varied , and they were 100 µg/ml for spectinomycin , 0 . 1 µg/ml for rifampin , 80 µg/ml for gentamicin , 4 µg/ml for chloramphenicol and 50 µg/ml for kanamycin . We opted to use antibiotic resistance gene cassettes against spectinomycin ( which also confers resistance to streptomycin [32] and chloramphenicol for assessing the growth of E . chaffeensis following the introduction of insertion mutations in the organism , as they were proven valuable in similar studies in other intracellular bacteria of the genera Anaplasma , Coxiella and Rickettsia [32] , [34] , [35] . A gentamicin resistance cassette was also used in this study , but as the results for this cassette were similar to the chloramphenicol acetyl transferase ( CAT ) gene cassette ( described below ) , the use of this cassette was not described . For creating targeted insertion mutations , several genomic regions spanning both intergenic and intragenic regions were selected . The genomic targets within protein coding sequences included genes Ech_0126 ( a hypothetical protein gene ) , Ech_1136 ( p28-Omp 14 gene ) and Ech_1143 ( p28-Omp 19 gene ) . The intergenic regions included the DNA segments spanning the genes Ech_0039 and Ech_0040 , Ech_0111 and Ech_0112 , and Ech_0251 and Ech_0252 . The hypothetical protein genes were selected because our RNA analysis ( not shown ) suggested that the bacterium did not transcribe from these genomic regions when propagated in DH82 macrophages . Ech_1136 is a major transcriptionally active gene encoding the p28-Omp 14 protein when the pathogen is grown in tick cells , whereas Ech_1143 gene is highly active in producing the p28-Omp 19 protein during its replication in vertebrate macrophages [15]–[18] . The underlying hypothesis for selecting these two differentially expressed genes is that disruption of a gene expressed exclusively in tick cells should permit normal growth of the organism in macrophages . Similarly , insertion mutations of genes expressed in macrophage environment should allow the pathogen to grow normally in tick cells . Two types of recombinant DNA segments were prepared targeting the gene Ech_0126; in the first version ( Rec I ) , a segment was prepared to introduce an antibiotic resistance gene to disrupt the gene . The recombination event results in the increase of a genome size by 0 . 93 kb . As the increase in genome size may cause polar effects influencing expression of genes surrounding the insertion site , a second type of recombinant segment was created to minimize the genome size increase in mutants . In this strategy ( Rec II ) , the length of the Ech_0126 segment was reduced to approximately equal the length of the antibiotic resistance cassette introduced . The homologous recombination schemes to achieve this are presented as a cartoon in supplemental Figure 1 ( Figure S1 ) . Independent of the recombination methods used , mutants were identified which persisted up to six days in culture and the presence of antibiotic had no impact on the mutants recovery; the mutants were assessed by Southern blot analysis of the PCR products generated with primers specific to the E . chaffeensis genomic region upstream of the insertion site and the inserted CAT gene sequence ( Figure S2 ) . Insertion junctions were verified by sequencing of the PCR products ( data not shown ) . Similar experiments were performed with gentamicin resistance cassettes and the mutants were detected similarly for six days . We reasoned that the lack of persistently growing mutants in culture may have resulted because the genomic region selected for the homologous recombination may be necessary for the bacterium's continued growth . Alternatively , the antibiotic resistance cassettes containing E . coli codon usage may not be optimal for protein expression in E . chaffeensis . We modified the codon usage for the CAT and gentamicin resistance gene cassettes to optimize their expression in E . chaffeensis and the mutational experiments were repeated . The targeted insertion mutants were detected similar to non-codon optimized constructs , but as in previous experiments the mutants survived only for up to eight days ( not shown ) . We expanded the targeted mutational experiments to six genomic sites by utilizing a mobile group II intron-based mutagenesis method ( the TargeTron method ) [25]–[28] . We opted to utilize this method as many studies in recent years reported its application for creating targeted mutations in several Gram-negative and Gram-positive bacteria [29]–[31] . The mobile group II intron was modified for insertions at six genomic locations of E . chaffeensis , including into the coding regions of Ech_0126 , Ech_1136 , and Ech_1143 and into intergenic regions of the genes Ech_0039 and Ech_0040 , Ech_0111 and Ech_0112 , and Ech_0251 and Ech_0252 . We inserted the pathogen's tuf promoter to promote expression of modified group II intron RNA and intron encoded protein in E . chaffeensis . The CAT gene cassette was also part of the modified mobile group II introns to confer resistance against chloramphenicol . Multiple assessments of insertions in all of the target regions led to the consistent identification of mutants when the mobile group II intron was targeted to the genes Ech_0126 , Ech_1143 and to the non-coding region between the genes Ech_0111 and Ech_0112 , but not in other three locations . Similar to the observations for the homologous recombination at Ech_0126 gene , the mutants persisted only for up to 8 days in culture . The presence of insertions was confirmed by Southern blot analysis of the PCR products generated with primers specific to insertion segment and to the surrounding genomic regions ( Figure S3 ) . Recent studies have demonstrated that Himar1 transposon mutagenesis can be reliably achieved in a rickettsial organism closely related to E . chaffeensis , i . e . , A . phagocytophilum [32] . In creating random mutations in E . chaffeensis , we utilized Himar1 transposon constructs similar to the ones used for mutagenesis of A . phagocytophilum [32]; the constructs prepared for this study represent single plasmids containing the Himar1 transposase gene and a transposon segment comprising of genes encoding mCherry or green fluorescent protein ( GFPuv ) co-expressed with the aad gene to confer resistance to spectinomycin and streptomycin flanked by inverted repeats recognized by the transposase . Three independent electroporation experiments were performed ( once with the mCherry plasmid and twice with the GFPuv plasmid ) using ISE6 tick cell-derived E . chaffeensis organisms; the organisms were then propagated in both ISE6 cells and DH82 cells . Antibiotic resistant organisms that persisted in culture for several months were identified using dual selection with spectinomycin and streptomycin ( 100 µg/ml each ) . The cultures also expressed mCherry or GFPuv ( Figure 1 ) . The transposon insertions in the mutants were verified by Southern blot analysis performed with an aad gene probe ( Figure 2 ) . Several restriction enzymes lacking recognition sites in the region that bound the DNA probe were used in the analysis . Several insertion sites were observed in the mCherry mutant DNA , as judged from the Southern blot signals ( both weak and strong signals ) . Similarly , multiple insertions were observed in the first GFPuv mutant DNA . In the second transformation experiment with the GFPuv plasmid , we recognized only one strong insertion segment . The intensity and the presence or absence of the hybridized DNA fragments in the mCherry and in the 1st GFPuv plasmid transformants were variable in DNA isolated from the organisms recovered from infected macrophages or tick cells ( Figure 2 , panel A ) . Similarly , hybridized fragments were variable in cultures when harvested at different times following electroporation ( data were presented for the cultures grown in macrophage cell line ) ( Figure 2 , panel B ) . To map the genomic locations of the transposon insertions , inverse PCRs and ST-PCRs ( semi-random , two-step PCR ) followed by sequence analysis of the PCR products were performed . We mapped five genomic locations of the mCherry transposon inserts and four genomic locations from the GFPuv insets ( three from the 1st and one from the 2nd GFPuv transformants ) ( Figure 3 ) . Six of the 9 transposon insertions were present within the non-coding regions of the genome; one of the insertions is close to the termination codon of the gene Ech_0230 ( just 18 base pairs 3′ to the termination codon ) . The remaining three insertions were present within the coding regions of three hypothetical protein encoding genes Ech_0379 , and Ech_0601 , and Ech_0660 . The insertions at all 9 genomic sites were confirmed by PCR and sequence analysis with primers designed to bind to the regions targeting the inserted DNA and to the surrounding genomic locations ( Figure 4 ) . Himar1 transposon insertions occur mostly once or twice per genome [36]–[38] . In this study , we observed multiple insertions in the mCherry transposon mutants and similarly for the first experiment performed with the GFPuv construct . We reasoned that the mutants represented heterogeneous mixture of organisms; this is consistent with varied signals in the Southern blots for the mutants isolated at different times in culture and for the observed differences in mutants recovered from macrophage and tick cell cultures ( Figure 2 ) . To test the hypothesis that the mutants represent a heterogeneous mixture , we performed serial dilution culture experiments with 10 fold dilutions of cell-free organisms mixed into uninfected macrophage cultures , and the infection in higher dilutions ( an estimated number of bacteria in the range of 0 . 1 to 10 per well ) was monitored for growth . We used Southern blotting ( not shown ) and PCR to target all 8 genomic insertions in cultures that tested positive for bacterial growth ( Figure 5 ) . The PCR products were observed for the clonal population only for the insertion downstream to the Ech_0284 ( lane 2c in Figure 5 ) from the mCherry mutant pool following the dilution cloning experiment . Similar efforts in recovering clonal populations from GFPuv mutant cultures resulted in the elimination of one of the three mutants from the pool ( mutant downstream to the Ech_0202 ) ( Lane 6c in Figure 5 ) Three of the 9 transposon mutants included insertions into the protein coding regions of three different hypothetical protein genes; Ech_0379 , Ech_0601 and Ech_0660 . To evaluate if the insertions resulted in the loss of gene activity from these genes , mRNA recovered from the wild-type and mutant organisms grown in culture was assessed by RT-PCR . All three genes are transcriptionally active in wild-type bacteria replicating in macrophages and two of the three genes are also active during growth in AAE2 tick cells ( Ech_0379 and Ech_0601 ) ( Figure 6 ) . Transcript levels for Ech_0379 were , however , low in the tick cell-derived bacteria , whereas Ech_0601 expression levels were similar in both macrophages and tick cells . Transposon insertions into the coding sequences of all three genes resulted in complete loss of gene expression ( Figure 6 ) without impacting growth in macrophage or tick cell cultures in vitro . As the insertion mutation near Ech_0230 is only 18 base pairs downstream from its coding sequence , we also assessed the impact of this mutation on the gene expression . This gene is expressed in wild-type organism grown in both the macrophage and tick cell cultures . The mutation , however , resulted in the loss of gene expression ( Figure 6 ) . We investigated whether one or more mutations into the coding or non-coding regions affected the pathogen's growth and persistence in a mammalian host or its acquisition by tick . A laboratory reared , white-tailed deer was inoculated intravenously with a mixed ( i . e . , not a clonal population ) mCherry transformant culture containing all five different insertion mutants . The presence of mutants in the inoculum was confirmed by Southern blot and insertion-specific PCRs ( not shown ) . Deer blood was monitored for infection by culture recovery of the organisms and PCR analysis targeting to all five insertion sites and for the presence of spectinomycin resistance gene for up to 63 days . With the exception of a few post-infection dates , the animal tested positive for the culture recoverable pathogen and insertion-specific , spectinomycin resistance gene ( aad ) ( Table 1 ) throughout the study period , suggesting persistence of the mutant organisms . The pathogens recovered from deer also expressed functional mCherry protein similar to the mutants in culture prior to infecting the animals ( data not shown ) . Evaluation of deer blood-derived total genomic DNA by insertion-specific PCRs targeting to all five genomic locations originally identified in the mCherry transformants resulted in the amplification of segments only from three intergenic locations ( Table 1 ) . Amplicons diagnostic for the insertion within the coding region of the gene Ech_0379 and one insertion into the non-coding region located 18 base pairs downstream from the Ech_0230 coding sequence were negative in all blood samples analyzed . Furthermore , the culture recovered organisms from blood samples evaluated by insertion-specific PCRs confirmed the presence of mutations only at the same three non-coding regions ( not shown ) . To confirm these results , we repeated the deer-infection experiment using mCherry transformants in two additional animals , and followed the infection for 49 days . As in the first experiment , mutants persisted in deer included only the same three intergenic insertions , whereas transformants with mutations into the coding region of Ech_0379 and the non-coding region mutation at the 3′ end of the Ech_0230 termination codon were not detected ( Table 2 ) . Nymphal A . americanum ticks were allowed to feed on a deer starting on day five post infection; fully engorged ticks that dropped off the animal were allowed to molt . Ten randomly collected molted adult ticks of both sexes were tested for E . chaffeensis transformant DNA by nested PCR targeting the spectinomycin resistance gene . Seven ticks that tested positive for the mutants were further assessed for the presence of mutations at all five genomic insertion sites ( Table 3 ) . The tick DNAs tested PCR positive for insertion mutations in two intergenic locations , and tested negative for the insertion within the coding region of the gene Ech_0379 and the non-coding insertion near Ech_0230 gene coding sequence , as observed from the PCR results for deer blood . The ticks were also negative for one other insertion located in the non-coding region between the genes Ech_0284 and Ech_0285 . The mutant pool infection experiments described above are similar to studies reported in the literature in identifying virulence associated genes by negative selection of several bacterial pathogens [39]–[41]; it will be a valuable tool for identifying additional genes essential for the E . chaffeensis infection in vivo . To further validate that the negative selection approach aids in identifying essential genes of E . chaffeensis , we used two clonal populations of mutants in another infection study in deer . One of these mutants had insertion within the coding region of the gene Ech_0660 and the second one has a mutation in the intergenic region located between the genes Ech_0284 and Ech_0285 . This clonal pair was selected because the Ech_0660 mutant was identified as having single insertion in the genome ( Figure 2 ) resulting in the loss of gene expression ( Figure 6 ) and the mutant in the intergenic region between Ech_0284 and Ech_0285 is one of the mutants that grew continuously in infected deer when inoculated as part of the pool of mutants ( described above ) . Blood samples collected from the infected animals for up to 20 days were assessed to monitor the mutants in circulation by performing nested PCR analysis ( Table 4 ) . The Ech_0660 infected animals tested negative for blood samples assessed for all post-infection days , whereas the infection with the intergenic mutant clone persisted in deer , as evidenced by the PCR positives on several post-infection days .
Ehrlichia species cause important diseases in people and animals , and understanding the pathogenesis and host-specific differences in gene expression of these pathogens are important goals . However , a major hurdle in advancing research in Ehrlichia species has been the lack of genetic tools , specifically , reliable methods for mutagenesis . This may largely be due to the fact that these organisms can survive only for a very short period of time outside a host cell [20] . Moreover , these phagocytotropic organisms do not appear to harbor plasmids [42] , and may not be amenable to plasmid transformation . Although preliminary evidence resulting from attempts to mutagenize E . chaffeensis have been discussed at scientific meetings [43] , [44] , stable mutants have not been realized until now . In this study , we took a methodical approach in demonstrating the feasibility of creating both targeted and random mutations in E . chaffeensis . Firstly , we evaluated various experimental conditions to identify protocols that work the best in transforming the organism with a plasmid ( results not discussed ) . Secondly , we assessed various antibiotics in support of identifying drugs that are useful in monitoring the mutants . We then tested the feasibility of creating targeted mutations by homologous recombination at a site which appeared to be transcriptionally silent . We subsequently expanded the targeted mutagenesis to six genomic locations using the modified mobile group II intron method , as recent research had demonstrated that it is an efficient method for creating targeted mutations in several Gram positive and Gram negative bacteria [28]–[31] . The method aided in the consistent identification of mutations in three targets assessed , including one located within a differentially expressed gene ( Ech_1143 ) that encodes an outer membrane protein , p28-Omp 19 . The mutants , however , survived in culture only for a short period of up to 8 days . We then evaluated transposon-based random mutagenesis and demonstrated that stable E . chaffeensis mutants could be generated in this way . Random mutagenesis resulted in multiple mutations at several genomic locations . The random mutants included within the coding regions of two differentially expressed genes ( Ech_0379 and Ech_0660 ) encoding for two different hypothetical proteins and one in the coding region of a constitutively expressed hypothetical protein gene ( Ech_0601 ) . Together , the experiments described in this study demonstrate that it is possible to create mutations within E . chaffeensis genes in both constitutively and differentially expressed genes , as well as in intergenic regions . The choice of mutagenesis methods does not seem to have any impact in creating mutations at a specific region of the genome . For example , we used three independent methods in creating targeted insertion mutations into the Ech_0126 gene ( two different approaches of homologous recombination and a TargeTron method ) . The use of all three methods aided in the identification of disruption mutations within this gene . It is not clear why the targeted mutations did not survive longer in culture; several possibilities should be considered: 1 ) it is possible that the genomic regions selected for insertion mutations may represent regions that are necessary for the organism's normal persistent growth in culture . 2 ) Alternatively , the E . chaffeensis promoter ( rpsl ) utilized in driving the expression from CAT gene may not be optimal in E . chaffeensis conferring resistance against the antibiotic concentration used . Although the rpsl promoter worked well in developing resistance in E . coli in driving the CAT gene expression ( not shown ) , its expression in E . chaffeensis may be an issue due to duplication of the rpsl promoter sequence in the genome . We are now investigating the use of heterologous promoters in targeted mutation experiments , such as the A . marginale tr promoter , as we already demonstrated its utility in the current study in the transposon mutagenesis in E . chaffeensis . Our efforts to create targeted mutations resulted in the establishment of mutants only in three out of six genomic targets evaluated . We reasoned that the targets selected for the three regions where mutants could not be detected might represent essential regions of the genome . One in particular , the Ech_1136 gene encodes for a protein ( p28-Omp14 ) that is differentially expressed with significant expression being observed in tick cells in vitro and in vivo [15]–[17] , [45] , [46] . The p28-Omp 14 is also a highly immunogenic protein; we reasoned that this gene product is an important component and its expression is critical for the organism's survival in a tick cell environment , but may be less critical for its growth in macrophages . The mutational analysis , however , suggested that the insertion mutation into this gene results in the complete inhibition of E . chaffeensis growth; this conclusion may infer that this gene product is necessary for the organism's growth in macrophages under in vitro conditions . This hypothesis is consistent with our recent studies suggesting that the protein from this gene is also expressed in macrophages [15] . Interestingly , targeted mutation in the predominantly macrophage-specific outer membrane protein gene of p28-Omp 19 ( Ech_1143 ) , a paralog of the p28-Omp 14 gene , appears to have no impact on the growth of the organism in macrophage cultures . Southern blot analysis of DNA recovered from transposon mutants harvested at different times during their growth in macrophage or tick cell cultures suggested that the mutants are a heterogeneous population . The Himar1 transposase mutations typically occur only at one or two locations in a genome [36]–[38] . To verify that the Himar transposon mutations in E . chaffeensis are similarly occurring at fewer genomic locations , we attempted to clone the mutants by performing a serial dilution cloning experiment . Establishing clonal populations from mutant pools is challenging and is a time consuming task for intracellular pathogens , such as E . chaffeensis . However , we were able to identify a single insertion containing culture from the mCherry mutant population and could eliminate one mutant from the GFPuv mutant pool . We mapped 9 random insertion mutation sites in the E . chaffeensis genome; six were located within non-coding regions and three were present within coding regions of three different transcriptionally active genes , and the mutations within the coding regions inhibited the mRNA production . Two of the active genes disrupted by the Himar1 insertion were differentially expressed genes [expressed highly ( Ech_0379 ) or exclusively ( Ech_0660 ) in macrophage cultures] . The complete inactivation of the genes caused by these mutations did not affect the growth of the organisms in either macrophages or tick cells . The nine genomic locations are the first to be identified as non-essential for E . chaffeensis growth in vitro in both vertebrate and tick cells; however , not all mutants grew in white-tailed deer ( the natural reservoir host for the pathogen [47] ) , and not all were acquired by its transmitting vector tick , A . americanum [47] , [48] . In particular , the transposon insertion into the coding region of the hypothetical protein encoding genes , Ech_0379 and Ech_0660 , and into the non-coding region near the 3′ end of Ech_0230 , prevented growth of the mutant organisms in deer . We were puzzled with the observation that the insertion mutation in a non-coding region also limited the growth of the organism in deer . One possibility is that insertions at certain non-coding regions of a genome may cause a polar effect influencing gene expression from surrounding genes leading to abolished growth in vivo , similar to the observations noted in the literature for other organisms [49] . Our RNA analysis , however , revealed that the mutation at the 3′ end of the ECH_0230 caused the inactivation of gene expression from this gene . The 3′ end mutation may have altered the transcript stability of Ech_0230 , possibly because it may be located within the DNA segment representing the transcriptional unit . We performed the infections in deer with pool of mutants in three separate animals using two independently cultured batches of organisms and the analysis also included 31 blood samples collected at various post-infection days ( 10 or 11 harvest points from each animal ) . All five mutants were represented in the culture pool used for infections . Independent of a blood sample analyzed , the only mutants consistently not detected in deer blood were those with the loss of gene expression ( in genes Ech_0230 and Ech_0379 ) . The in vivo survived mutants in deer , therefore , represent only those for which the transcriptional activity was not negatively affected in genes near the insertion mutations . More importantly , mutations causing the loss of gene expression correlated very well with their failed growth in deer or their non-acquisition by ticks during blood feeding . Together , these data suggest that the elimination of some mutants in vivo from a pool of mutants is the result of the E . chaffeensis growth inhibition because of the loss of gene function probably impacting the survival of the pathogen in vivo . The protocols outlined in this study , beginning from creating mutations and assessing their ability to grow or not to grow in a host ( Figure 7 ) , will serve as a useful tool in screening the genome of E . chaffeensis to identify additional genes contributing to the pathogenesis . The in vivo infection assessment with pools of mutants to identify virulence associated genes in bacterial pathogens is well-documented in the literature [39]–[41] . Our study , however , is the first to apply similar strategy for an obligate intra cellular rickettsial pathogen and it is likely valuable in initiating similar studies in other related pathogens . The method may also be applicable for other obligate intracellular pathogens , such as Chlamydia for which genetic manipulation methods are still at infancy [50] . Establishing clonal populations from mutant pools is a challenging task for intracellular pathogens , such as E . chaffeensis . We , however , were able to prepare two clonal populations from the mutants described in this study; one having mutation causing the loss of gene expression ( Ech_0660 ) and the second one having a mutation into a noncoding region ( between the genes Ech_0284 and Ech_0285 ) with no impact of gene expression in genes 5′ and 3′ to the insertion mutation ( data not shown ) . Infections with the two clones in deer further supported the conclusions drawn from the infection experiments using pool of mutants . The Ech_0660 mutant causing the gene expression inactivation also failed to grow in deer , while the intergenic mutant persisted well . The lack of growth of the subset of mutants from a pool of mutants in deer , therefore , is not a random event because the only mutants that failed to grow in deer are those impacting the gene expression of the organism . Ticks also did not acquire E . chaffeensis mutants containing gene function disruptions . One other mutant with insertion into the non-coding region between the genes Ech_0284 and Ech_0285 also did not grow in ticks . Failure of ticks to acquire the gene disruption mutants is likely due to their inability to productively infect deer during the time of tick feeding . The third mutant ( mutant # 2 ) that infected deer but was not acquired by ticks , may represent an insertion mutation at a genomic region required for E . chaffeensis growth in ticks . The alternate possibility is that the specific mutant organisms were not in circulation in deer during the period of tick feeding . Specific functions of the disrupted genomic regions remain to be determined . The gene Ech_0230 encodes for a putative membrane protein and Ech_0379 and Ech_0660 are identified as genes encoding for hypothetical proteins [38] ( GenBank # CP000236 . 1 ) . The gene product for the Ech_0660 contains a putative conserved domain having significant homology to phage proteins involved in the phage capsid assembly ( observed in the BLAST search analysis ) , suggesting that it may represent a membrane expressed protein of E . chaffeensis . Protein database homology search on the translated coding sequence of Ech_0379 revealed a significant homology for a 46 amino acid long N-terminus domain with a membrane-bound , monovalent cation ( Na+ ) /H+ antiporter protein subunit of two Corynebacterium species; C . ammoniagenes and C . casei ( Figure 8 ) . Ech_0379 gene homologs are also conserved in other Ehrlichia species , e . g . , in E . ruminantium , and included similar functional domain conservation to antiporters . Moreover in E . ruminantium , its homolog is classified as an integral membrane protein ( Figure 8 ) . It is unclear if the Ech_0379 gene product indeed represents a protein with antiporter function . Most bacterial genomes contain several genes encoding for antiporters that exchange Na+ and/or K+ for H+ from outside the cell [51] . On the contrary , intracellular pathogenic bacteria usually possess at most one antiporter . The E . chaffeensis genome , however , contains five homologs with obvious homology to antiporters ( GenBank # CP000236 . 1 ) . As the antiporters aid bacteria in meeting challenges of high or fluctuating pH , salt , temperature or osmolarity [51] , they may represent an important set of proteins for an organism's intracellular survival in vivo . E . chaffeensis resides in a phagosome environment and most likely depends heavily on antiporters for its pH balance in a vertebrate host . Although we do not have any data supporting that Ech_0379 gene encodes an antiporter , the research reported in this study clearly demonstrates that it is an essential gene for the organism . Together , the predicted gene products of all three genes required for the pathogen's growth in vivo ( Ech_0230 , Ech_0379 and Ech_0660 ) appear to be associated with a membrane structure of E . chaffeensis . This is the first study describing the targeted mutational methods for Anaplasmataceae pathogens . Similarly , this is the first in reporting the utility of transposon-based random mutagenesis in Ehrlichia species . Evidence was also presented for the first time in showing mutants impacting the growth of the organism in its natural host . Additional experimental manipulations are necessary for optimizing the protocols for recovering stable targeted mutants . A combination of the use of a targeted mutagenesis coupled with a method in expressing antisense RNA can aid in evaluating the importance of additional differentially expressed genes to the pathogen's adaptation to dual hosts and in causing pathogenesis . In particular , it is now possible in creating targeted mutations at a genomic location that does not impact the growth of the organism , such as the ones we identified in this study by random mutagenesis . Targeted insertions at such locations to synthesize complementary RNA of a gene of interest with an inducible promoter may be used to knockdown gene expression [52]–[54] . Creation of a large library of clones containing mutations in various genomic locations and assessing their importance to the pathogen's growth in vivo is feasible for bacteria having small genomes , such as E . chaffeensis . The molecular tools described in the current study , therefore , should aid in advancing our understanding of the role specific genes and genomic regions have in supporting the dual host life cycle of this and other related tick-borne pathogens .
E . chaffeensis ( Arkansas isolate ) was continuously cultivated in the canine macrophage cell line ( DH82 ) essentially as described earlier [55] . The ISE6 and AAE2 cell lines originating from Ixodes scapularis and A . americanum ticks , respectively [56] were also used to cultivate E . chaffeensis by following the protocols reported earlier [16] . Detailed protocols for propagating the organisms were followed as described earlier [57] . E . chaffeensis was cultivated in T25 flasks having approximately 80% confluent DH82 cells . When the infectivity reached approximately 50% , cultures were detached from the flask using sterile glass beads and 0 . 5 ml each of the culture suspension was transferred to a 24 well plate . Spectinomycin , rifampin , gentamicin , chloramphenicol , kanamycin and ampicillin were tested for their ability to inhibit E . chaffeensis growth . The concentrations of the antibiotics tested were as follows: spectinomycin at 0 , 10 , 50 , 100 , 150 and 200 µg/ml; rifampin at 0 , 0 . 1 , 0 . 2 and 0 . 5 µg/ml; gentamicin at 0 , 10 , 20 , 40 , 60 and 80 µg/ml; chloramphenicol at 0 , 2 , 4 , 6 , 8 and 10 µg/ml; Kanamycin at 0 , 10 , 20 , 30 , 40 , 50 and 60 µg/ml . The assays were performed in triplicate wells for each antibiotic concentration assessed . E . chaffeensis infection levels in DH82 cultures were monitored once every three days for 20 days or longer by microscopic evaluation [52] . Various primers used in preparing the mutational constructs and for assessing the insertions as well as for analyzing gene expression are listed in Supplemental Table ( Table S1 ) . The primers were custom ordered from a commercial vendor ( Integrated DNA Technologies , Inc . , Coralville IA ) . Based on RT-PCR analysis and cDNA microarray ( Cheng et al . unpublished results and Kuriakose et al . [58] ) , some of the E . chaffeensis genes appeared to be non-expressed or expressed at low levels during growth of the organisms in macrophage or in tick cell cultures . Some of these genes were selected for creating targeted mutations by homologous recombination or a mobile group II intron based mutagenesis approach commonly referred to as the TargeTron method . Ech_0126 , which appeared to be a transcriptionally silent gene , was chosen to build the homologous recombination constructs by two different formats ( Rec I and Rec II ) . The Rec I construct was built by amplifying a 2 . 7 kb size fragment of the E . chaffeensis ( Arkansas isolate ) Ech_0126 gene spanning the genomic region from 113 , 771 to 116 , 518 ( GenBank # CP000236 . 1 ) . To promote directional cloning into the plasmid , primers were designed to contain restriction sites for Not I at the 5′ end and for Hind III at the 3′ end . The fragment was initially cloned into the PCR 2 . 1 Topo vector ( Life Technologies , CA ) . The recombinant plasmid DNA was then digested with the Bstz17 I enzyme ( which cuts at a site located1 . 07 kb from the 5′ end of the Ech_0126 insert ) , and used for inserting the antibiotic resistance cassette , CAT gene or the gentamicin resistance gene . [The antibiotic resistance genes were amplified from the plasmid pDEST10 ( Life Technologies , CA ) . ] To drive gene expression , the rpsl promoter , generated by PCR from the E . chaffeensis genome , was inserted upstream of the antibiotic resistance cassettes . The functionality of the rpsl promoter in driving the expression of the antibiotic resistance cassette genes was verified in E . coli . The rpsl promoter represents one of the highly active promoters in the E . chaffeensis genome ( our unpublished results ) , possibly because it is responsible for driving the expression of several genes ( GenBank # CP000236 . 1 ) . A homolog of this promoter was also used earlier to drive expression of foreign genes in a related rickettsia , Rickettsia prowazekii [23] , [24] . To create the Rec II construct , a 1 . 25 kb size 5′ fragment of the Ech_0126 gene ( E . chaffeensis genome sequence coordinates: 113 , 771–115 , 024 ) and a 1 . 49 kb fragment at the 3′ end of the Ech_0126 gene ( genome coordinates: 115 , 755–117 , 238 ) were amplified using E . chaffeensis genomic DNA as the template . PCR primers were designed to include a Not I site at the 5′ end and a Spe I site at the 3′ end for the 5′ end fragment , and Spe I site at the 5′ end and a Hind III site at the 3′ end for the 3′ end fragment to facilitate directional cloning into the plasmid vector . The PCR products were cloned into the PCR2 . 1 Topo vector by using the TA cloning method . The above described antibiotic resistance cassettes , including the rpsl promoter , were then cloned into the Spe I site . The inserts containing the E . chaffeensis Ech_0126 gene sequences and the antibiotic resistance cassette segments from the plasmid constructs were amplified using a proof reading enzyme , high fidelity platinum Taq DNA polymerase ( Lifetechnologies , CA ) to obtain linear fragments for use in homologous recombination experiments . The genes Ech_0126 , Ech_1136 , and Ech_1143 and the intergenic regions between the genes Ech_0039 and Ech_0040 , Ech_0111 and Ech_0112 and Ech_0251 and Ech_0252 were chosen for creating insertion mutations by employing the modified mobile group II intron-based TargeTron mutagenesis strategy ( Sigma-Aldrich , St . Louis , MO ) . The TargeTron plasmid constructs were prepared by following the manufacturer's instructions . In particular , we used the web-based software program offered by the vendor in identifying sequences having the highest probability in introducing the modified mobile group II intron at the above identified E . chaffeensis genomic regions . Primer sets were designed to target each genomic site and used to modify the group II intron segment present in the plasmid construct , PACD4KC . In support of the expression of the group II intron in E . chaffeensis , the organism's tuf promoter ( promoter for E . chaffeensis translation elongation factor gene , Ech_0407 ) was generated by PCR and cloned upstream of the modified group II intron segment by inserting at the Hind III restriction site . The activity of the tuf promoter was verified in E . coli for driving the expression of a promoterless β-galactosidase gene prior to its use in the E . chaffeensis experiments . To aid in selecting mutants at various genomic sites , the antibiotic resistance cassette , CAT gene , driven by the E . chaffeensis rpsl promoter was inserted at the Mlu II restriction enzyme site within the group II intron gene which replaced the built-in kanamycin resistance cassette . Himar1 A7 constructs used in this study were similar to the ones described earlier in [32] with minor modifications . Specifically , a single plasmid construct containing the Himar1 transposase gene driven by the Anaplasma marginale transcriptional regulator ( Am-tr ) promoter , and the transposon comprising a fluorescent protein marker gene and the spectinomycin/streptomycin antibiotic resistance gene ( aad ) , also driven by the Am-tr promoter , was created . The transposon was flanked by the inverted repeats recognized by the transposase . Two different variations of the plasmids encoded either the green fluorescent protein ( GFPuv ) or the mCherry gene cloned upstream of the antibiotic resistance protein coding sequence . The plasmids are named pCis GFPuv-SS Himar A7 and pCis mCherry-SS Himar A7 , respectively . For targeted mutagenesis assays , E . chaffeensis infected DH82 cultures at about 80% infectivity were harvested from a confluent T25 flask , centrifuged at 2000×g for 10 min to collect the cell pellet which was then resuspended in 10 ml of 0 . 3 M sucrose solution . The solution was passed through 27 . 5 gauge needle 5 times to lyse the host cells . The cell lysate was then filtered through 2 . 7 and 1 . 6 µm filters ( Millipore , Billerica , MA ) and the solution was centrifuged at 15 , 500×g for 10 min to pellet the host cell-free organisms and the pellets were washed twice with 10 ml of ice-clod 0 . 3 M sucrose and resuspended in 50 µl of 0 . 3 M sucrose solution for use in transformation . Ehrlichia organisms are quantitated by performing the real-time PCR assay as we described earlier [59] . Although the number varies each time , typically we observe approximately 100 bacteria per each infected host cell . For random mutagenesis using the Himar1 constructs , 3 . 5 ml of medium was carefully removed from one fully infected 5 ml ISE6 culture , and the cells were resuspended in the remaining medium . The suspension was transferred to a microfuge tube containing 0 . 2 ml of silicon carbide ( #1 course rock tumbling grit , Loretone Inc . , Mukilteo , WA ) , and vortexed for 30 sec at high speed . The supernatant was passed through a 2 µm pore-size filter ( Whatman Ltd . , Piscataway , NJ ) , and bacteria were collected by centrifugation at 11 , 000×g at 4°C for 5 min . Bacteria were washed twice in 0 . 3 M sucrose , and kept on ice between washes . Aliquots of purified E . chaffeensis ( ∼5×108 per construct to be used ) were resuspended in 50 µl of cold 0 . 3 M sucrose containing 1 µg of plasmid DNA and kept on ice for 15 min before electroporation ( see below ) . Six µg of purified linear PCR fragment from Rec I or Rec II constructs for homologous recombination or 6 µg of plasmid DNA for TargeTron or 1 µg of Himar1 plasmid DNA were mixed with 50 µl of host cell free Ehrlichia ( containing about 2×109 organisms for targeted mutagenesis or ∼5×108 for transposon mutagensis ) for each electroporation , transferred to 1 mm gap electroporation cuvettes , and incubated on ice for 15 min [32] . [Plasmid DNAs were prepared using a Maxiprep plasmid DNA isolation kit by following the manufacturer's instructions ( Qiagen , CA ) . ] E . chaffeensis organisms were electroporated at 2 , 000 volts , 25 µF and 400 Ω setting . The mixture was then combined with 0 . 5 ml of fetal bovine serum and 1 ml of DH82 or ISE6 cell suspension containing about 1×106 cells . The sample was centrifuged at 5000×g for 5 min , incubated at room temperature for 15 min and then transferred into a sterile T25 flask with a confluent monolayer of DH82 or ISE6 cells . Cultures were incubated at 30°C overnight and then transferred to a 34°C incubator for ISE6 cells or 37°C incubator for DH82 cells for the continuous growth of the organisms . Three days later , an appropriate antibiotic with a desired concentration was added to the culture media in support of selecting the mutants containing the resistance gene against a specific antibiotic . The experiments with each targeted mutation construct were repeated at least three independent times . Himar1 mutants were selected in the presence of 100 µg/ml of spectinomycin and 100 µg/ml of streptomycin . The culture media containing antibiotics were replaced twice a week for DH82 cells and three times per week for ISE6 cells . When infectivity reached to 80% or higher , cell free Ehrlichia were prepared for inoculating a new flask of uninfected host cells with medium containing antibiotics . This procedure was repeated until all wild-type bacteria were eliminated . The presence of insertion mutations was monitored for 60 days or longer . Genomic DNAs from the transformant flasks were extracted using the Puregene genomic DNA isolation kit as described by the manufacturer ( Qiagen Inc . , Valencia , CA ) . For targeted mutation analysis , PCR fragments amplified using a genome specific primer paired with an insertion specific primer from the 5′ end or 3′ end of integration region were amplified and the products were resolved on a 0 . 9% agarose gel , transferred to a nylon membrane and probed with an insertion specific probe . Once predicted size amplicons were identified , the products were purified and sequenced using the CEQ 8000 genetic analyzer ( Beckman Coulter , CA ) . Sequences generated from this experiment were compared with the genomic region to identify the junctions . For Himar1 transposon mutant analysis , 100 ng of genomic DNA recovered from the cultured organisms was digested with Spe I/Nde I , Bsrg I or EcoR V/Bgl II restriction enzymes for 3 hours at 37°C , resolved on a 0 . 9% agarose gel for about 6 hours at 60 volts , transferred to a nylon membrane by capillary transfer and then probed with a 32P labeled insertion specific probe at 68°C overnight by following standard molecular biology protocols [60] . The membrane was then exposed to an X-ray film to capture the radioactive signals from the hybridized DNA fragments . The genomic locations of the insertions were identified by inverse PCR or ST-PCR methods as described earlier [60] , [61] . Briefly , for inverse PCR , 1 µg of genomic DNA from the mutant Ehrlichia cells were digested with Hind III , the restricted fragments were resolved on an agarose gel , specific genomic fragments were then purified and diluted in nuclease-free water; the DNA was ligated with T4 DNA ligase to favor the intramolecular ligation in a 400 µl volume . The ligated products were then used for inverse PCR [60] using insertion-specific primers designed to face away from each other on the genomic DNA strands . For ST-PCR , the initial PCR was performed using a random primer having a GATAT at the 3′ end and a degenerate sequence at the 5′ end paired with an insertion specific primer . The first PCR product was then purified and diluted to serve as the template in the nested PCR using a target specific nested primer and the second primer designed to prime to the degenerate sequence inserted in the first round of PCR by following the protocols described in [61] . A primer set was designed for use in the amplification of a segment from the genomic insertion sites; in this primer set , one primer was designed to bind to either 5′ or 3′ end of the insertion specific sequence and the second primer was designed to bind to the genomic region either upstream or downstream of the insertion site . The primers and the genomic DNAs isolated from the mutant cultures were used in the amplification assays and specific product amplification was verified by subjecting the PCR products to sequence analysis using a CEQ 8000 genetic analyzer ( Beckman Coulter , CA ) . Total RNA from wild-type E . chaffeensis grown in DH82 or AAE2 cultures and Himar1 transformants grown in DH82 cultures was isolated by using Tri-reagent method following the manufacturer's instructions ( Sigma-Aldrich , St . Louis , MO ) . Primers targeting to Ech_0379 , Ech_0601 or Ech_0660 were designed for use in RT-PCR analysis . Total RNA was treated with RQ1 DNase at 37°C for 60 min to remove any genomic DNA contamination . For each set of RT-PCRs , one tube with reverse transcriptase , one tube without reverse transcriptase , one using DNA as the template and one with no template were included . For verifying RNA expression , the presence or absence of specific products in the assays containing RNA and reverse transcriptase was assessed and compared to the product generated from genomic DNA . Animal experiments with deer were performed to comply with the Public Health Service ( PHS ) Policy on the Humane Care and Use of Laboratory Animals , the US Department of Agriculture's ( USDA ) Animal Welfare Act & Regulations ( 9CFR Chapter 1 , 2 . 31 ) , and with the prior approval of the Oklahoma State University ( OSU ) Institutional Animal Care and Use Committee ( IACUC ) . Veterinary care for the animals was overseen by a University Veterinarian . At the end of each experiment , deer were euthanized and euthanasia was performed also in accordance with the university IACUC recommendations that are consistent with the recommendations of the Panel on Euthanasia of the American Veterinary Medical Association . Three day-old white-tailed deer fawns purchased from a breeder were reared in a tick free laboratory environment until the age of 3–5 months prior to utilizing them for experimental infection studies . Deer rearing and experimental infections were performed at Oklahoma State University ( OSU ) as per the guidelines outlined in the protocol . Deer were housed in a natural environment until utilized for infection studies . Fawns were moved to an Animal Biosafety Level facility for all experiments . Every effort is made to attend to the comfort and wellbeing of the deer . Food and water are provided ad libitum . During infection of the deer with E . chaffeensis , deer are housed in specially designed rubber-matted round enclosures twelve feet in diameter . Although confined to an enclosure , the deer were able to move around freely and were not restrained . Infected deer were monitored daily for signs of illness ( decreased activity , decreased appetite , ruffed hair , nasal discharge , sneezing/coughing , etc . ) . Blood and serum samples were collected a maximum of two times per week and no more than 10 ml of blood was collected at each blood draw . We observed very mild to no symptoms associated with E . chaffeensis infections . At the end of each experiment , deer were euthanized as indicated above . Tissue harvests and animal disposal was done at the Oklahoma Animal Disease Diagnostic Laboratory . E . chaffeensis mCherry transformant cultures at about 80–90% infectivity in T75 flask were harvested ( about 15 ml ) , centrifuged at 15 , 000 g for 10 min at 4°C , supernatant was discarded and the culture was resuspended in 15 ml of 1×phosphate buffered saline ( PBS ) . The washing steps were repeated twice and the final cell pellet was suspended in 7 . 5 ml of PBS to concentrate infected DH82 cells to about 2×106 per ml ( the estimated concentration of Ehrlichia organisms in the cells was approximately 2×108 per ml ) . One ml each of the cell suspension was used for intravenous injections into a deer . The experiments were performed two independent times using freshly prepared cultures; the first experiment included one animal and the second experiment included two animals , all of which were inoculated with mCherry mutants . Deer infections were also performed similarly using the clonal population of mutants with mutation into the coding region of Ech_0660 ( three deer ) or intergenic region mutation into the non-coding region between the genes Ech_0284 and Ech_0285 ( two deer ) . About 3 ml of each of deer blood was collected in sterile EDTA tubes on day zero ( prior to infection ) and on days 3 , 5 , 6 , 8 , 10 , 14 , 18 , 21 , 28 , 32 , 35 , 39 , 42 , 46 , 49 , 52 , 56 , 59 , and 63 post-inoculation for the first experiment . Blood samples for the second infection were also collected on day zero and on 3 , 7 , 10 , 13 , 17 , 20 , 23 , 30 , 37 and 49 days post inoculation . The third infection blood samples were collected on day zero and on 3 , 6 , 9 , 13 , 16 , and 20 days post inoculation . The blood samples were stored at 4°C until use ( maximum of three days ) . Blood samples were spun at 3 , 000 rpm in a Clay Adams Sero-fuge ( Becton Dickinson , Sparks , MD ) for 5 min . Plasma was removed and about 1 ml of buffy coat each was transferred to a 15 ml sterile Falcon centrifuge tube containing 10 ml RBC lysis buffer ( 155 mM NH4Cl , 10 mM KHCO3 and 0 . 1 mM EDTA ) and mixed several times until complete lysis of erythrocytes . The samples were then centrifuged at 5 , 000 g for 5 min and the supernatants were discarded . The buffy coat pellet from each sample was resuspended in 300 µl of 1×PBS . To assess for infection with E . chaffeensis , 100 µl each of the cell suspensions was transferred into a well of 12-well sterile culture plate containing 0 . 9 ml of DH82 cells having about 80% confluency . The cultures were grown by following the detailed culture protocols reported earlier [57] and infection was monitored twice a week by examining the Hema3 stained cytospin slides for up to 8–10 weeks to determine if a sample was positive or negative for the organisms . Contents of positive infection were transferred to a T25 flask and allowed to grow until infection reached about 80% and were used for genomic DNA isolation or for liquid nitrogen storage . One hundred µl each of the buffy coats from deer blood were also used for isolating total genomic DNA by using the Wizard SV Genomic DNA purification kit as per the manufacturer's instructions ( Promega , Madison , WI ) ; purified DNA from each sample was stored in 100 µl of buffer containing 10 mM Tris-HCl and 1 mM EDTA ( pH 8 . 0 ) ( TE buffer ) . The DNAs were used to assess E . chaffeensis infection status by performing semi-nested PCR targeting the insertion specific spectinomycin resistance gene ( aad ) ( primers for this experiment are listed in Table S1 ) . Briefly , 2 µl of genomic DNA from deer blood was used for the first round PCR in a 25 µl reaction volume using Platinum Taq DNA polymerase as per the manufacturer's instructions ( Life Technologies , Grand Island , NY ) . The PCRs were performed in a GenAmp9700 instrument ( Applied Biosystems , Foster City , CA ) with the following temperature cycles: 94°C for 4 min , followed by 35 cycles of 94°C for 30 s , 52°C for 30 s , and 72°C for 1 min and 1 cycle of 72°C for 3 min . The nested PCR was done using the same PCR conditions as the first round PCR and the templates for the second round included 2 µl of 1∶100 diluted products from the first PCR and with nested PCR primer set . Samples that were positive for the aad gene were also evaluated by nested PCRs targeting the transposon insertions and flanking genomic regions in the five mCherry mutants defined ( primers for this experiment are listed in Table S1 ) . Nymphal A . americanum ticks were obtained from the National Tick Research and Education Resource ( NTRER ) at Oklahoma State University . Laboratory-reared ticks were propagated following the published protocols of Patrick and Hair [62]; and in accordance with the approval from the IACUC and as per the guidelines outlined in the approved protocol . About 300 nymphal A . americanum ticks were placed on infected deer starting day 5 post inoculation . Engorged ticks were collected after blood meal ( typically in about 7 days ) and stored at room temperature and 96% humidity chamber and permitted to molt to adults ( took approximately 45–50 days ) . Genomic DNA was isolated from individual adult ticks by using the Wizard SV Genomic DNA purification kit as outlined above . Final purified DNAs were resuspended in 100 µl each of TE buffer . Two µl each was then used for nested PCR analysis targeting to the aad gene or targeting to the transposon insertion regions as described above . | The tick-transmitted bacterium , Ehrlichia chaffeensis , causes human monocytic ehrlichiosis , an acute febrile illness that can progress to a fatal outcome . This and other related pathogens have evolved to establish infections in vertebrate and tick hosts for completing their lifecycle . Our recent studies suggest that the pathogen's differential gene expression during growth in ticks and mammals is a major contributor for its dual host adaptation . However , the importance of the pathogen phenotype differences is best understood if we have methods to knock down protein expression from one or more genes . Creating mutations in obligate intracellular pathogens remain a challenge due to their limited survival in the extracellular environment . Here , we present evidence for multiple insertion mutations in the E . chaffeensis genome . Three of the nine mutations in the genome inhibiting gene expression prevented infection of deer , the natural host for the pathogen . This is the first study demonstrating the feasibility of creating mutations in an Ehrlichia species; and directly linking specific regions of the genome to in vivo infection . Methods described here allow for studies to define genes important for infectivity and ability to cause disease , and are equally important for initiating similar studies in other related emerging zoonotic pathogens . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"microbial",
"mutation",
"vector",
"biology",
"microbial",
"pathogens",
"microbial",
"control",
"genetics",
"and",
"genomics",
"biology",
"genomics",
"microbiology",
"ticks",
"host-pathogen",
"interaction",
"bacterial",
"pathogens",
"microbial",
"growth",
"and",
"development"
] | 2013 | Targeted and Random Mutagenesis of Ehrlichia chaffeensis for the Identification of Genes Required for In vivo Infection |
Urease is a metalloenzyme essential for the survival of Helicobacter pylori in acidic gastric environment . Maturation of urease involves carbamylation of Lys219 and insertion of two nickel ions at its active site . This process requires GTP hydrolysis and the formation of a preactivation complex consisting of apo-urease and urease accessory proteins UreF , UreH , and UreG . UreF and UreH form a complex to recruit UreG , which is a SIMIBI class GTPase , to the preactivation complex . We report here the crystal structure of the UreG/UreF/UreH complex , which illustrates how UreF and UreH facilitate dimerization of UreG , and assembles its metal binding site by juxtaposing two invariant Cys66-Pro67-His68 metal binding motif at the interface to form the ( UreG/UreF/UreH ) 2 complex . Interaction studies revealed that addition of nickel and GTP to the UreG/UreF/UreH complex releases a UreG dimer that binds a nickel ion at the dimeric interface . Substitution of Cys66 and His68 with alanine abolishes the formation of the nickel-charged UreG dimer . This nickel-charged UreG dimer can activate urease in vitro in the presence of the UreF/UreH complex . Static light scattering and atomic absorption spectroscopy measurements demonstrated that the nickel-charged UreG dimer , upon GTP hydrolysis , reverts to its monomeric form and releases nickel to urease . Based on our results , we propose a mechanism on how urease accessory proteins facilitate maturation of urease .
More than a quarter of the cellular proteins are metalloproteins [1] . Homeostatic regulation on the incorporation of specific metal ions to these metalloproteins is essential to life . Metal ions that occupy the top of the Irving-Williams series [2] , such as nickel , copper , and zinc , form stable complexes with proteins . Therefore , these metals must be tightly regulated in the cell to be kept out of other proteins that require less competitive metals to function . Specificity of metal incorporation into metalloprotein is conferred by specific protein–protein interactions with metallochaperones . Urease is a nickel containing metalloenzyme that hydrolyses urea into ammonia . It is a bacterial virulence factor that enables the survival of bacteria through the use of urea as the sole nitrogen source under nitrition limiting conditions [3] . Moreover , its activity enables Helicobacter pylori to survive under strongly acidic conditions of the human stomach by neutralizing gastric acid with the ammonia released [4] . Maturation of urease , a process that involves the delivery of two nickel ions into the carbamylated active site of the apo-enzyme , offers a paradigm to study metallochaperone-driven incorporation of metal into metalloenzyme . Four urease accessory proteins—UreE , UreF , UreG , and UreH ( UreH is an ortholog of UreD found in other species ) —participate in the regulation of nickel delivery and urease maturation [5] . These genes were first identified in Klebsiella aerogenes using deletion and complementation approaches [6] and their homologs were later identified in H . pylori in a similar fashion [7] . In vitro pull-down assay showed that apo-urease forms a series of complexes with K . aerogenes urease accessory proteins—namely , UreD/urease [8] , UreF/UreD/urease [9] , and UreG/UreF/UreD/urease [10] . The interactions between urease and its accessory proteins were further supported by yeast-two-hybrid [11]–[13] and tandem affinity purification . It was hypothesized that the UreG/UreF/UreH complex is responsible for delivery of nickel ions into the urease active site and the formation of the preactivation complex is a critical step involved in urease maturation . Truncation study on K . aerogenes UreE–UreF fusion protein demonstrated that the C-terminal region of UreF is essential for interaction with other urease accessory proteins [14] . We have recently solved the crystal structure of UreF/UreH complex and demonstrated that UreH induces conformational changes in UreF to facilitate the recruitment of UreG [15] . Substitutions on the conserved residues at the C-terminal tail of UreF disrupt the formation of UreG/UreF/UreH complex [15] . Failure to recruit UreG to form UreG/UreF/UreH complex also resulted in abolishment of urease maturation [15] , [16] . UreG is a SIMIBI ( after signal recognition particle , MinD , and BioD ) class GTPase involved in the regulation of nickel delivery [17] . Biological functions of SIMIBI GTPases are frequently regulated by dimerization [18] . Its GTPase activity is essential for urease maturation as either the substitution of GTPase P-loop motif or the use of nonhydrolyzable GTP analog during urease activation resulted in inactive urease [10] , [19] , [20] . UreG contains an invariant metal binding Cys-Pro-His motif [21] , [22] , substitution of which abolishes urease maturation [22] . It has been shown that UreG can interact with UreE , which was hypothesized as a nickel carrier that supplies nickel to urease during the maturation process [23]–[25] . How urease accessory proteins facilitate maturation of urease is not well understood . It is not known how GTP hydrolysis of UreG is coupled to urease activation , and why the recruitment of UreG to the urease preactivation complex is essential for urease maturation . Here , we report the structure of the UreG/UreF/UreH complex . Together with our biochemical study , we have revealed how urease accessory proteins couple GTP hydrolysis to nickel delivery , which may provide a paradigm for other metal-delivering NTPases .
To gain insight into the urease maturation process , we determined the structure of the UreG/UreF/UreH complex in its GDP-bound state to 2 . 35 Å resolution using X-ray crystallography ( Table S1 ) . Coordinates of the structure have been deposited in the Protein Data Bank ( PDB Code: 4HI0; the coordinates were communicated to Prof . R . P . Hausinger prior to publication and our structure was mentioned in a recent review article by Hausinger et al . [26] ) . The asymmetric unit contains two copies of each of UreF , UreH , and UreG , related by 2-fold symmetry , forming a dimer of heterotrimers ( Figure 1A ) . UreF and UreH in the UreG/UreF/UreH complex have essentially the same conformation as previously reported in the UreF/UreH complex ( Figure S1 , Cα r . m . s . d . 0 . 742 Å ) . The UreG/UreF/UreH complex structure represents the first known structure of UreG ( Figure 2A ) . The topology of UreG is characteristic of SIMIBI class GTPases containing the canonical G1 to G5 motif for guanine nucleotide recognition ( Figure S2 ) . Two GDP ligands are sandwiched between two UreG protomers ( Figure 1A ) . UreG contains an invariant Cys66-Pro67-His68 metal binding motif ( Figure S3 ) [27] , substitution of which abolishes the metal binding properties of UreG [22] , [27] . This metal binding motif is found at the dimer interface between two UreG protomers ( Figure 2A ) . Cys66 and His68 from each protomer are arranged in a symmetric formation that could potentially bind a metal ion at the interface ( Figure 2A ) . The crystal structure also showed that UreG interacts strictly with UreF only , leading to an overall structure resembling that of an inverted-T shape ( Figure 1A ) . Each UreG protomer makes extensive contacts with both protomers of UreF . The contact surface area between one UreG protomer ( chain E ) and UreF chain A is 1 , 664 Å2 and with chain C is 1 , 004 Å2 , respectively . The interaction between UreG and UreF is mostly mediated via specific hydrogen bonds and salt bridges ( Table S2 ) , with the interacting residues distributed evenly in a ring-shaped pattern on the UreG dimer surface contacting UreF ( Figure 2B ) . These interacting residues are positioned in close proximity to the GTPase switch I and II ( G2 and G3 ) motifs and the metal binding motifs of UreG ( Figures 1B and 2A ) . Moreover , each UreG protomer is in intimate contact with the residues on the F-tail loops of both protomers of UreF ( Figure 1B ) . This observation is consistent with previous structural and mutagenesis analyses , which suggest that binding of UreH induces the ordering of the F-tail loop structure required for the recruitment of UreG [15] , [16] . As each UreG protomer interacts with both protomers of UreF ( Figure 1B ) , we hypothesized that dimerization of UreF/UreH to form the ( UreF/UreH ) 2 complex is prerequisite to UreG recruitment . To test this hypothesis , we created a UreF variant ( R179A/Y183D ) ( Figure 3A ) that disrupted the homodimerization of UreF/UreH ( Figure 3B ) . We showed that the substitutions resulted in a dimerization-deficient UreF ( R179A/Y183D ) /UreH complex that failed to recruit UreG ( Figure 3C ) . Next , we test if the UreF variant can activate urease in vivo ( Figure 3D ) . Whereas the wild-type pHpA2H plasmid gave an activity of 0 . 31±0 . 02 µmol NH3 mg−1 min−1 , pHpA2H-ureF ( R179A/Y183D ) mutant plasmid gave activity similar to that of the negative control of pHpAB , containing only the urease structural genes ( Figure 3D ) . Taken together , our results suggest that homodimerization of the UreF/UreH complex is essential for the recruitment of two copies of UreG to form the ( UreG/UreF/UreH ) 2 complex and urease maturation . In an attempt to prepare the UreG/UreF/UreH complex in its GTP-bound state , we have accidentally discovered that UreG has a tendency to dissociate from the UreG/UreF/UreH complex upon addition of GTP . We performed GST pull-down experiments to test the interaction between UreG and UreF/UreH in the presence of GDP or GTP . Our results showed that UreG partially dissociated from the GST-UreF/UreH complex in the presence of GTP ( Figure 4A , lanes 5 and 6 ) . On the other hand , complete dissociation of UreG was observed upon addition of both GTP and nickel ( Figure 4A , lanes 11 and 12 ) . Similar observation was obtained when GTP and zinc was added ( Figure S4 , lanes 5 and 6 ) . These observations suggest that GTP binding promotes dissociation of UreG from the UreF/UreH complex , the process of which is enhanced by addition of nickel or zinc . That the addition of both GTP and nickel dissociated UreG from the UreF/UreH complex was further supported by size exclusion chromatography/static light scattering ( SEC/SLS ) analysis . The UreG/UreF/UreH complex was eluted as a single peak at 12 . 5 ml , with a molecular weight of 156 . 7±0 . 5 kDa , which is consistent with the expected molecular weight of a heterohexameric ( UreG/UreF/UreH ) 2 complex ( Figure 4B , injection 1 ) . Addition of GTP and nickel ions to the UreG/UreF/UreH complex resulted in elution of two peaks at 12 . 9 ml and 15 . 1 ml with molecular weights consistent with that of the UreF/UreH complex and UreG dimer ( Figure 4B , injection 2 ) . SDS-PAGE analysis of the eluted fractions confirmed the identity of the proteins eluted in the corresponding peaks ( Figure 4C ) . The dissociation of UreG dimer from UreF/UreH complex upon the addition of GTP and nickel prompted us to systematically characterize the oligomerization state of UreG in the presence of GDP , GTP , and/or nickel ions . SEC/SLS analysis showed that regardless of the addition of guanine nucleotides , UreG eluted in a monomeric form ( molecular weight ∼24 kDa ) in the absence of nickel ( Figure 5A , injections 1 to 3 ) . In agreement with a previous study , we found that UreG remains a monomer in the presence of nickel ( Figure 5A , injection 4 ) [21] . UreG showed some tendency to dimerize in the presence of nickel and GDP as indicated by the presence of a minor peak with molecular weight of 44 . 5±1 . 8 kDa ( Figure 5A , injection 5 ) . In the presence of both GTP and nickel ion , however , UreG eluted as a stable dimer with a molecular weight of 46 . 8±0 . 2 kDa ( Figure 5A , injection 6 ) . We further accessed the nickel content of UreG dimers by collecting the protein sample eluted from the SEC/SLS system and measuring its nickel content using atomic absorption spectroscopy ( AAS ) . Our result shows that the protein sample contained 1 . 2±0 . 1 bound nickel per UreG dimer ( Figure 5C ) . To demonstrate that the invariant Cys66-Pro67-His68 metal binding motif of UreG is responsible for its GTP-dependent dimerization and nickel binding properties , we repeated the above experiments using UreG ( C66A ) and UreG ( H68A ) variants . We found that while wild-type UreG dimerizes in the presence of GTP and nickel , both UreG variants remained as monomers ( Figure 5B ) . Both UreG variants also have significantly reduced nickel chelating ability as compared to the wild-type UreG ( Figure 5C ) . Thus , our data suggest that the UreG dimer binds a nickel ion with the invariant Cys66-Pro67-His68 motif at the dimeric interface . As observed in the crystal structure of the UreG/UreF/UreH complex , the metal binding motif of UreG is situated at the interface between two UreG protomers ( Figure 2A ) . Moreover , SEC/SLS analysis of UreG above suggested that the oligomerization state of UreG is related to its guanine nucleotide state ( Figure 5A ) . These observations give rise to an attractive hypothesis that the transitioning of UreG between its monomeric and dimeric forms enables it to dynamically assemble and disassemble the metal binding site to deliver nickel to its target urease . We first tested if the nickel-charged UreG dimer can revert to monomer upon hydrolysis of bound GTP . Since bicarbonate is a known cofactor required for urease activation [28] , [29] , we speculated that bicarbonate stimulates GTP hydrolysis by UreG . We found that after incubation in the absence of bicarbonate at 37°C for 3 h , UreG mostly remained in the dimeric state ( Figure 6A , upper panel ) . In contrast , all the UreG became monomeric in the presence of bicarbonate ( Figure 6A , upper panel ) . We then measured the phosphate released during the incubation period using a malachite green-based assay . We found that each nickel-charged UreG dimer , when stimulated by bicarbonate , released 2 . 0±0 . 4 phosphate ( Figure 6B ) , whereas only a negligible amount of phosphate was released when GTP and nickel are incubated with bicarbonate in the absence of UreG ( Figure S5 ) . In comparison , only 0 . 3±0 . 1 phosphate was released in the absence of bicarbonate ( Figure 6B ) . When the experiment was repeated using UreG dimers prepared with GTPγS , which is a nonhydrolyzable analog of GTP , we found that UreG remained mostly in its dimeric form regardless of the presence of bicarbonate ( Figure 6A , lower panel ) . Our results suggest that GTP hydrolysis stimulated by bicarbonate dissociates UreG dimer into monomer . In addition , we found that neither acetate , formate , nor sulfate is capable of stimulating phosphate release from UreG dimer ( Figure S6 ) . Therefore , stimulation of UreG GTPase activity appears to be specific to bicarbonate . We then tested whether the bound nickel of UreG dimer is released upon GTP hydrolysis . We compared the nickel content of the dimeric and monomeric forms of UreG using AAS . Our results showed that after 3 h incubation in the presence of bicarbonate , each GTP-containing UreG dimer released 1 . 0±0 . 1 nickel ion . In comparison , less than 0 . 10 nickel was released from UreG dimer containing GTPγS ( Figure 6C ) . Taken together , monomeric UreG dimerizes in the presence of both GTP and nickel ions . Stimulated by bicarbonate , UreG hydrolyzes GTP to GDP and returns to its monomeric form while releasing nickel ions in the process ( Figure 6D ) . To test the metal specificity of GTP-dependent dimerization of UreG , we compared the dimerization behavior of UreG with nickel or zinc using SEC/SLS . UreG was mixed with nickel or zinc sulfate in the absence/presence of guanine nucleotides ( Figure S7A ) . The protein sample was then loaded to a gel filtration column and eluted with a buffer without metal ions . Our results showed that UreG only formed a stable dimer with the addition of both nickel and GTP in the protein sample . It was reported that zinc can induce dimerization of UreG [21] , [30] . However , in their experiments , they included 10 µM zinc in the gel filtration buffer . Consistent with their finding , we found that UreG dimerized irrespective to the guanine nucleotide state when zinc was included in the gel filtration buffer ( Figure S7B ) , but remained monomeric when the metal was absent from the buffer ( Figure S7A ) . These findings suggest that zinc-induced UreG dimer is stable only when a constant amount of free zinc ion is in solution . In contrast , UreG was able to maintain a stable dimeric form with GTP even when nickel was not present in the gel filtration buffer ( Figure S7A ) . Given the limited pool of free metal ions in the cytosol under cellular conditions , we deem that GTP-dependent dimerization of UreG is specific for nickel . Considering both structural and biochemical data , we showed that dimerization of the UreF/UreH complex provides a platform to recruit two copies of UreG in a GDP-bound state to form a heterohexameric complex ( UreG/UreF/UreH ) 2 . Addition of nickel and GTP dissociates the complex into the ( UreF/UreH ) 2 complex and a nickel-charged UreG dimer ( Figure 4 ) . We then questioned whether the nickel-charged UreG dimer is biologically active and capable of activating urease . To this end , we performed an in vitro urease activation assay using purified proteins , without external sources of nickel or GTP other than those from the UreG dimers . We found that the nickel-charged UreG dimer can activate urease only in the presence of the UreF/UreH complex ( Figure 7A ) . In contrast , without the addition of UreF/UreH , the nickel-charged UreG dimer alone was not capable of urease activation ( Figure 7A ) . Moreover , only GTP-bound UreG dimers could activate urease but not those with GTPγS ( Figure 7A ) , suggesting that GTP hydrolysis is a requirement for urease activation . Consistent with the observation that bicarbonate can stimulate GTPase activity of UreG , we also found that urease can be activated only in the presence of bicarbonate ( Figure 7B ) . Since bicarbonate can also stimulate the release of nickel from the UreG dimer ( Figure 6 ) , we tested if free nickel at a concentration equivalent to that of UreG dimer ( 20 µM ) can activate urease . Our results showed that the addition of nickel resulted in only ∼18% urease activation in the absence of UreG ( Figure 7C ) . This observation suggests that the bound nickel in the UreG dimer is important in urease activation , and it is likely that GTP hydrolysis of UreG and nickel release have to couple with the UreF/UreH/urease complex to achieve efficient urease activation . Nickel-charged UreG dimer can activate urease in vitro in the presence of UreF/UreH ( Figure 7A ) , suggesting nickel ions were transferred from the UreG dimer to urease . We next tested whether the nickel-charged UreG dimer can interact with the UreF/UreH/urease complex . Pull-down assay showed that the nickel-charged UreG dimer does not interact with the GST-UreF/UreH complex ( Figure 7D ) . This indicates that the nickel-charged UreG dimer does not reassociate with the UreF/UreH complex . Interestingly , we found that the nickel-charged UreG dimer can interact with the GST-UreF/UreH/urease complex ( Figure 7D ) . Presumably , conformational changes resulting from the formation of the UreF/UreH/urease complex are necessary to accommodate the UreG dimer . Nevertheless , our result implies that nickel is transferred from UreG to urease through the complex formation with UreF/UreH and urease .
Urease is a metalloenzyme whose catalytic activity is crucially dependent on the placement of the proper metal ion in the active site . This is achieved in vivo through the use of four urease accessory proteins in a GTP-hydrolysis-dependent process [10] . It is well known that these urease accessory proteins form a series of complexes with urease [5] , leading to the formation of the urease preactivation complex consisting of UreF , UreH , UreG , and urease . Crystal structures of the UreF and UreF/UreH complex provided a structural rationale for the sequence of assembling components in the preactivation complex [15] , [31] . Using chemical cross-linking , mutagenesis , and SAXS experiments , it has been shown that UreF/UreH binding induces conformational changes on urease [32] , [33] . UreF has been suggested to increase the fidelity of nickel incorporation in urease [16] . However , less is understood on how nickel is inserted into urease . In this study , we have determined the crystal structure of the UreG/UreF/UreH complex . Static light scattering measurements established that while apo-UreG is monomeric in solution ( Figure 5A ) , the complex exists in a dimeric ( UreG/UreF/UreH ) 2 formation similar to that observed in the crystal structure ( Figure 4B ) . Together , the structure shows that ( UreF/UreH ) 2 provides a scaffold for UreG dimerization . In support of this , it has been previously observed that UreF variants with substitution of conserved residues at the UreF-UreG interface abolished both recruitment of UreG and maturation of urease [15] , [16] . Moreover , UreF/UreH dimerization is essential to urease maturation , as suggested by our mutagenesis study showing that UreF with substitutions to disrupt the homodimerization of UreF/UreH complex also failed to recruit UreG and abolishes urease maturation ( Figure 3 ) . Taken together , we conclude that UreF/UreH-assisted dimerization of UreG is a prerequisite for the production of active urease . Our structure further indicates that dimerization of UreG brings together the invariant Cys-Pro-His metal binding motifs on each UreG protomer , suggesting that nickel is chelated at the dimeric interface of UreG ( Figure 2A ) . Next , we showed that nickel can induce GTP-dependent dimerization of UreG ( Figure 5A ) and that a stable nickel-charged UreG dimer can be purified . The role of the Cys-Pro-His motif in binding metal was confirmed by the observation that UreG variants with either C66A or H68A substitutions abolished nickel-induced GTP-dependent dimerization of UreG ( Figure 5B ) . Furthermore , equivalent substitutions on K . aerogenes UreG abolished urease maturation [22] . Together , this suggests UreG dimerization results in the chelation of a nickel ion that is essential to urease maturation . Further experimentation with the UreG dimer showed that nickel is released from this dimer upon GTP hydrolysis ( Figure 6B and 6C ) , which accounts for the previously shown requirement of GTP hydrolysis for urease activation [10] . Our results also provide insights into how UreG couples GTP hydrolysis to the delivery of nickel to urease [10] . We showed that the nickel-charged UreG dimer can form a complex with UreF/UreH and urease , which upon stimulation of GTPase activity with bicarbonate resulted in activated urease ( Figure 7 ) . How the UreG dimer interacts with UreF/UreH/urease is not known . Since the UreG dimer only interacts with the UreF/UreH/urease complex but not UreF/UreH , we speculate conformational changes resulting from the formation of the UreF/UreH/urease complex are necessary to accommodate the UreG dimer . We additionally note that in the GST pull-down experiment demonstrating an interaction between UreG dimer and UreF/UreH/urease complex , UreB ( α chain of urease ) appears to be substoichiometric in comparison to UreA ( β chain of urease ) ( Figure 7D ) . This suggests the possibility that some UreB may have dissociated from UreA when interacting with the UreG/UreF/UreH complex and direct interaction occurs between UreG/UreF/UreH and UreA but not UreB . Nevertheless , our results demonstrated that the nickel-charged UreG dimer is able to activate UreF/UreH-bound apo-urease in a GTP-hydrolysis-dependent process ( Figure 8A ) . It is interesting to note that the invariant metal binding motif Cys-Pro-His of UreG is located next to the switch I and II regions ( Figure S8 ) . The canonical loaded-spring switch mechanism universal to many GTPases such as Ras and Rho [34] involves conformational changes in switch I and II regions . Given that the Cys-Pro-His motif is in close contact with residues in the switch II region ( Figure S8 ) , it can be anticipated that GTP hydrolysis can induce conformational changes at the Cys-Pro-His motif located at the dimeric interface , altering the nickel chelation environment and/or dissociation of UreG into monomer , thus leading to the release of nickel to the urease . How nickel is transferred to UreG in vivo is currently not known . It has been postulated that UreE is the nickel carrier protein and a likely source of nickel in urease maturation [35]–[37] . Structures of UreE with a nickel bound at the interface between two UreE protomers was reported [23] , [38] and characterized using X-ray absorption spectroscopy [38] . Moreover , UreE and UreG interaction was detected using yeast-two-hybrid and tandem affinity purification approaches [11]–[13] , [39] . Affinity pull-downs showed that K . aerogenes UreG and UreE form a complex in the presence of nickel [22] . The strength of the interaction between UreG and UreE was measured using surface plasmon resonance and biolayer interferometry techniques [40] . These observations all support the idea of a nickel transfer between UreE and UreG via protein–protein interaction . As proposed by Hausinger et al . [26] , nickel-carrying UreE may directly transfer nickel to the UreG/UreF/UreH/urease complex ( Figure 8B , red arrows ) . This model is supported by the observation that UreE can interact with UreG/UreF/UreD/urease [22] . It is intriguing that in the crystal structure of UreG/UreF/UreH , the metal binding motif of UreG is buried and there are no obvious channels available in the UreG/UreF/UreH structure for the passage of nickel ions . Given that UreF/UreH can induce large conformational changes in urease [32] , [33] , conformational changes in the UreG/UreF/UreH/urease complex may occur to allow the transfer of nickel from UreE to the metal binding site of UreG . Alternatively , the nickel transfer from UreE to UreG occurs before the formation of the UreG/UreF/UreH/urease complex ( Figure 8B , blue arrows ) . Our results suggest that GTP can weaken the interaction between UreG and UreF/UreH ( Figure 4A ) . Since the switch I and II regions of UreG are located near its UreF-interacting residues ( Figure 2A ) , it is conceivable that binding of GTP promotes dissociation of UreG , which then receives a nickel from UreE , perhaps in the form of a ( UreE/UreG ) 2 complex as suggested in a modeling attempt by Bellucci et al . [27] . The resulting nickel-charged UreG dimer can then form complex with UreF/UreH and urease . In either case , the preactivation complex , involving nickel-charged UreG , UreF/UreH , and urease , delivers nickel to the urease active site upon stimulation of GTPase activity by bicarbonate ( Figure 8A ) . As with other GTPase-based molecular switches , it is imperative for UreG to be controlled precisely on the timing of nickel release to function effectively . Previous modeling studies argued that UreF may serve as a GTPase-activating protein ( GAP ) based on sequence homology [41] , [42] . Biagi et al . generated a computational model of the UreE/UreG/UreF/UreH complex [42] . This model depicted UreG in an inverted orientation as compared to what was observed in the crystal structure of the UreG/UreF/UreH complex . The Biagi model suggests that the nucleotide binding site of UreG is in contact with UreF while the metal binding site is away from the UreF interface . However , this is inconsistent with our experimentally determined crystal structure of the UreG/UreF/UreH complex , which showed that the metal binding site is pointing towards the UreF while the nucleotide binding site is far away from the UreF/UreG interface ( Figure 1 ) . Moreover , GTPase activity measurements in K . aerogenes found no evidence for GAP activity by UreF [16] . Taken together , both pieces of experimental data indicate that it is unlikely for UreF to serve as GAP by supplying an arginine finger for the stimulation of GTP hydrolysis as in the case of Ras [43] . Rather interestingly , we found that it is bicarbonate ion , which is a known cofactor involved in urease maturation [28] , that triggers the GTPase activity of UreG , leading to the release of nickel ( Figure 6B and 6C ) and activation of urease ( Figure 7B ) . Similarly , Boer et al . observed that GTPase activity of UreG in K . aerogenes leads to phosphate release in the presence of bicarbonate under in vitro conditions [16] . We have shown that the ability to stimulate GTPase activity of UreG is specific for bicarbonate ( Figure S6 ) . That other carboxylate-group-containing anions , such as acetate and formate , did not stimulate GTPase activity suggests that it is unlikely for the bicarbonate ion to serve as a surrogate of the carbamylated Lys219 in the urease active site to signal UreG for the release of nickel . Alternatively , it is possible that GTPase activity of UreG is directed for the synthesis of carboxylphosphate as suggested [10] and thereby coupling the carbamylation of active site Lys219 to nickel release from UreG . Interestingly , it has been reported that in the absence of guanine nucleotide , UreG has a stronger binding affinity to zinc than nickel ( Kd 0 . 33 µM versus 10 µM ) [21] . Here , our results imply that the nickel affinity of UreG is modulated by guanine-nucleotide binding state . In the absence of free metal ion in the gel filtration buffer , Zn-induced dimer readily dissociated into monomer , while the GTP/nickel-charged UreG dimer remained dimeric ( Figure S7 ) . The observation implies that the GTP-bound state of UreG has a stronger binding affinity to nickel than zinc . As shown in the structure of UreG/UreF/UreH , the Cys-Pro-His motif of UreG is located next to the switch I and II regions ( Figure S8 ) , and conformational changes upon GTP binding may alter the metal chelating environment to favor nickel binding . It makes biological sense to have a large difference in nickel binding affinity between GDP and GTP bound state of UreG because it provides a mechanism to release the bound nickel upon GTP hydrolysis . As lucidly argued in a review by Waldron and Robinson , it is the specific protein–protein interactions that drive the delivery of appropriate metal ion to the target metalloproteins [44] . Here , we have shown how urease accessory proteins form a specific complex and interplay with each other to couple GTP hydrolysis to deliver nickel into urease . It is important to recognize that UreG belongs to a growing family of G-proteins regulated by homodimerization [18] . Two other well-characterized nickel-delivering NTPases , namely HypB and CooC1 , share strikingly similar properties with UreG . HypB is a close relative of UreG responsible for delivering nickel into hydrogenases . Similar to UreG , it exhibits a varying degree of dimerization in the presence of guanine nucleotides , but achieves complete dimerization in the presence of both GTP and nickel [45] . Crystal structures of HypB in its apo form [46] and GTP/zinc bound form [47] have been reported , showing two invariant cysteine residues located at the dimer interface are responsible for metal chelation . Substitutions that abolished GTP-dependent dimerization of HypB also weaken hydrogenase maturation [46] , [48] , indicating dimerization is essential to the functioning of HypB . CooC1 is an ATPase responsible for delivery of nickel into carbon monoxide dehydrogenases and a distant relative of UreG in the SIMIBI class NTPases . The presence of nickel , ADP , or ATP induces dimerization of CooC1 [49] . Crystal structure of CooC1 shows that a conserved Cys-X-Cys motif at the dimer interface is responsible for metal chelation [50] . Given the similarity shared among UreG , HypB , and CooC1 , we believe that mechanism of nickel delivery described for UreG in this study may represent a general strategy for nickel delivery to other metalloenzymes .
The UreF/UreH complex was purified as described previously [15] . The UreF ( R179A/Y183A ) /UreH variant ( Figure 3B ) was purified following the same method as used for the wild-type protein . The expression vector of UreG was constructed by subcloning the ureG gene into an in-house pHisSUMO vector , which contains the coding sequence of a HisSUMO tag on a pRSETA vector ( Invitrogen ) . UreG was expressed as a HisSUMO tagged fusion protein using transformed E . coli . HisSUMO-UreG expressing bacterial cells were lysed via sonication in buffer A ( 20 mM Tris/HCl , pH 7 . 5 , 200 mM NaCl , 1 mM TCEP , and 40 mM imidazole ) and loaded onto a 5 ml HisTrap column ( GE Healthcare ) equilibrated with buffer A . After washing with 10 column volumes of buffer A , HisSUMO-UreG was eluted with 300 mM imidazole in buffer A . HisSUMO tag was cleaved using small ubiquitin-like modifier protease SENP1C and separated from UreG by a second pass through the HisTrap column . UreG was further purified by size exclusion chromatography using HiLoad 26/60 Superdex 75 column ( GE Healthcare ) in buffer A without imidazole . Purified UreG was dialyzed again with 5 mM EDTA to remove any bound metal before further experimentation . The UreG/UreF/UreH complex was prepared by mixing the UreF/UreH complex with 2-fold molar excess of UreG . The UreG/UreF/UreH complex was then isolated from excess UreG by size exclusion chromatography using HiLoad 26/60 Superdex 200 column ( GE Healthcare ) . The expression vector for H . pylori urease was constructed by subcloning ureA and ureB into pRSF-duet vector . Urease was expressed using transformed E . coli . Bacterial cells containing urease were lysed via sonication in buffer B ( 20 mM Tris pH 7 . 5 and 1 mM TCEP ) and loaded onto a 5 ml Q Sepharose column ( GE Healthcare ) equilibrated with buffer B . After extensive washing using buffer B , urease was eluted using a linear 0–500 mM sodium chloride gradient . Pulled fractions were further purified by size exclusion chromatography using HiLoad 26/60 Superdex 200 column ( GE Healthcare ) in buffer A without imidazole . For experiments involving the use of nickel-charged UreG dimers ( Figures 6 , 7 , S6 , and S7 ) , the UreG dimer was purified by incubating apo-UreG at 3 mg/ml on ice with 1 mM GTP , 2 mM magnesium sulfate , and 0 . 5 mM nickel sulfate for 15 min . The nickel-charged UreG dimer formed was then separated from excess GTP and nickel using a Sephadex G25 desalting column . The UreG/UreF/UreH complex was concentrated to 20 mg/ml for crystallization . Just before crystallization , 5 mM GDP , 10 mM magnesium sulfate , and 5 mM aluminum fluoride was added to the UreG/UreF/UreH complex . Crystals of the UreG/UreF/UreH complex was grown using sitting drop vapor diffusion method in 1 . 5 M ammonium sulfate and 100 mM sodium acetate pH 5 . 0 at 16°C . Crystals were harvested after 2 wk and cryoprotected by soaking in mother liquor containing 20% glycerol before flash freezing with liquid nitrogen for data collection . X-ray diffraction data were collected on beamline I-04 at a wavelength of 0 . 9795 Å at the Diamond Light Source ( Oxfordshire , UK ) . Diffraction data were integrated and scaled using XDS [51] . Initial phases were obtained using molecular replacement with PHASER [52] . Structures of the UreF/UreH complex ( PDB code: 3SF5 ) and HypB ( PDB code: 2HF9 ) devoid of the GTPγS ligand were used as molecular replacement search models . The resulting model was further improved by iterative rounds of refinement using PHENIX . REFINE [53] with noncrystallographic symmetry ( NCS ) restrains and manual model building with COOT [54] . After initial rounds of refinement , clear positive electron density was identified in the Fo-Fc map in the guanine nucleotide binding pocket of UreG , in which GDP was modeled ( Figure S9A ) . Translation-libration-screw ( TLS ) refinement procedure was applied in the final rounds of refinement . 2Fo-Fc map shows that clear electron density for the metal binding motif of UreG was observed ( Figure S9B ) . The Molprobity [55] validated final model has 97 . 84% and 2 . 16% in Ramachandran favored and allowed regions , with no outlier residues . To detect interaction of the UreF/UreH complex with UreG ( Figure 3C ) , bacterial cells expressing UreG and those expressing HisGST-UreF/UreH or HisGST-UreF ( R179A/Y183D ) /UreH complexes were mixed and sonicated together . Cleared cell lysate was loaded onto GST SpinTrap column ( GE Healthcare ) and incubated for 30 min at 4°C . The column was then washed 5 times with 500 µl of buffer C ( 20 mM Tris pH 7 . 5 , 200 mM sodium chloride , 1 mM TCEP ) , followed by elution using buffer C with 10 mM glutathione . Eluted protein was analyzed using 15% SDS-PAGE . To study the interaction between UreF/UreH and UreG in the presence of guanine nucleotide and/or nickel/zinc ions ( Figures 4A and S4 ) , purified GST-UreF/UreH and excess UreG was mixed and loaded onto GST SpinTrap column ( GE Healthcare ) . After incubating for 30 min at 4°C and washing away unbound proteins , the immobilized GST-UreG/UreF/UreH complex was incubated with 1 mM GTP or GDP , 2 mM magnesium sulfate , and/or 0 . 5 mM nickel sulfate for 30 min at 4°C . Column was then washed 5 times with 500 µl of buffer C , followed by elution using buffer C with 10 mM glutathione . The wash fraction and eluted fraction was collected and analyzed using 15% SDS-PAGE . To study the interaction between nickel-charged UreG dimer , UreF/UreH , and UreF/UreH/urease ( Figure 7D ) , purified GST-UreF/UreH or GST-UreF/UreH mixed with overexpressed urease lysate was loaded onto 5 ml GSTrap column ( GE Healthcare ) . After extensive washing with buffer C , apo-UreG or nickel-charged UreG dimer was injected into the column and incubated for 15 min . Column was then washed again with buffer C to remove unbound UreG and finally eluted using buffer C with 10 mM glutathione . For the study of dimerization-deficient UreF variant ( Figure 3 ) , 100 µl of 50 µM purified UreF/UreH complex or its mutants were injected into a Superdex 200 analytical gel filtration column pre-equilibrated with phosphate-buffered saline . For the study of nickel/guanine nucleotide–dependent dissociation of UreG from UreF/UreH ( Figure 4B ) and dimerization of UreG ( Figure 5A and 5B ) , 100 µl of 40 µM UreG or UreG/UreF/UreH complex pre-incubated with different combinations of 1 mM GTP/GDP , 2 mM magnesium sulfate , and 0 . 5 mM nickel sulfate was injected into the same column . For the study of bicarbonate-induced change in oligomerization state of nickel-charged UreG ( Figure 6A ) , 75 µM of nickel-charged UreG with or without prior incubation of 100 mM bicarbonate was injected into the same column . For testing the metal specificity of UreG dimerization ( Figure S7 ) , experimental conditions were chosen to closely match that used by Zambelli et al . [21] to make our results comparable . We pre-incubated 100 µM of UreG with 1 mM magnesium sulfate , 100 µM nickel sulfate , or zinc sulfate in the presence/absence of 0 . 5 mM GDP or GTP . The protein samples were injected into a Superdex 200 analytical gel filtration column pre-equilibrated the gel filtration buffer containing 20 mM Tris pH 8 . 0 and 150 mM NaCl . We included 10 µM of nickel or zinc in the gel filtration buffer in selected injections . In all cases , protein eluted off the gel filtration column was fed into a downstream miniDawn light scattering detector and an Optilab DSP refractometer ( Wyatt Technologies ) . Data were analyzed using ASTRA software provided by the manufacturer ( Wyatt Technologies ) , and the molecular weights measured along with the fitting errors are reported . UreG injected into the Superdex 200 analytical gel filtration/static light scattering system was eluted into a 4 ml fraction . Nitric acid was added to the eluted UreG to a final concentration of 1% . Nickel concentration was measured using Hitachi Z-2700 polarized Zeeman atomic absorption spectrometer with graphite furnace . Data were analyzed using software provided by the manufacturer . Nickel concentration was determined by comparing measurements against a standard curve of nickel known concentration ( Figures 5C and 6C ) . We incubated 200 µl of 75 µM of nickel-charged UreG in the absence or presence of 100 mM sodium bicarbonate for 3 h at 37°C . Phosphate released from UreG GTPase activity was measured using a colorimetric assay method based on malachite green [56] . Measurements were compared against a standard curve prepared using known amounts of phosphate to determine the phosphate concentration . Phosphate released by UreG dimer was derived from the difference between the amount of phosphate in solution before and after the incubation period ( Figures 6B , S5 , and S6 ) . For the study of dimerization-deficient UreF variant , a cell lysate-based urease activity assay was used ( Figure 3D ) . pHpA2H vector was constructed by cloning the entire urease operon into a pRSETA vector . pHpA2H mutant plasmids were constructed using Quikchange site-directed mutagenesis protocol ( Strategene ) . pHpA2H plasmids carrying mutations on either ureF gene were transformed into E . coli cells . Urease activity assay was performed as described [15] . For the study of investigating the ability of nickel-charged UreG in urease activation , an in vitro urease activity assay using purified proteins was used ( Figure 7 ) . The standard buffer for in vitro activation of urease consists of 20 mM HEPES pH 7 . 5 , 150 mM sodium chloride , 1 mM TCEP , and 100 mM sodium bicarbonate . To test the activation of urease , 20 µM of each of purified urease and UreF/UreH complex was incubated with 20 µM nickel-charged UreG dimers for 2 h at 37°C . Urease activity was then determined by incubating the enzyme with 25 mM urea for 30 min at 37°C and then measuring the ammonia released using a phenol/hypochlorite reaction as described [57] . | Catalytic activities of many important enzymes depend upon metal cofactors . Ensuring each enzyme acquires the proper type of metal cofactor is essential to life . One such example is urease , which is a nickel containing metalloenzyme catalyzing the hydrolysis of urea to ammonia . The survival of Helicobacter pylori , a stomach ulcer–causing pathogen , in the human stomach depends on the ammonia released to neutralize gastric acid . In this study , we revealed the detail mechanism of how urease accessory proteins UreF , UreH , and UreG cooperate to couple GTP hydrolysis to deliver nickel to urease . UreF/UreH complex interacts with two molecules of GTPase UreG and assembles a metal binding site located at the interface between two UreG molecules . Nickel can induce GTP-dependent dimerization of UreG . This nickel-carrying UreG dimer together with UreF , UreH , and urease assemble into a protein complex . Upon stimulation of UreG GTPase activity by bicarbonate , UreG hydrolyses GTP and releases nickel into urease . Other nickel-delivering NTPases share similar properties with UreG; therefore , the nickel delivery mechanism described here is likely universally shared among these proteins . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [] | 2013 | Structure of UreG/UreF/UreH Complex Reveals How Urease Accessory Proteins Facilitate Maturation of Helicobacter pylori Urease |
The neuropeptide Pigment Dispersing Factor ( PDF ) is essential for normal circadian function in Drosophila . It synchronizes the phases of M pacemakers , while in E pacemakers it decelerates their cycling and supports their amplitude . The PDF receptor ( PDF-R ) is present in both M and subsets of E cells . Activation of PDF-R stimulates cAMP increases in vitro and in M cells in vivo . The present study asks: What is the identity of downstream signaling components that are associated with PDF receptor in specific circadian pacemaker neurons ? Using live imaging of intact fly brains and transgenic RNAi , we show that adenylate cyclase AC3 underlies PDF signaling in M cells . Genetic disruptions of AC3 specifically disrupt PDF responses: they do not affect other Gs-coupled GPCR signaling in M cells , they can be rescued , and they do not represent developmental alterations . Knockdown of the Drosophila AKAP-like scaffolding protein Nervy also reduces PDF responses . Flies with AC3 alterations show behavioral syndromes consistent with known roles of M pacemakers as mediated by PDF . Surprisingly , disruption of AC3 does not alter PDF responses in E cells—the PDF-R ( + ) LNd . Within M pacemakers , PDF-R couples preferentially to a single AC , but PDF-R association with a different AC ( s ) is needed to explain PDF signaling in the E pacemakers . Thus critical pathways of circadian synchronization are mediated by highly specific second messenger components . These findings support a hypothesis that PDF signaling components within target cells are sequestered into “circadian signalosomes , ” whose compositions differ between E and M pacemaker cell types .
The importance of biological rhythms in the anticipation and response to daily environmental changes is underscored by their conservation throughout nature . In eukaryotes , these rhythms are generated by a set of core clock genes that contrive to produce interlocked feedback loops . Both mammalian and Drosophila circadian rhythms are controlled by diverse groups of pacemaker neurons that express these core clock genes and proteins . In Drosophila these rhythms are required in ∼150 neurons , which can be subdivided into six bilateral anatomically distinct groups [1] . There appear to be two classes of pacemaker neuron in the fly brain that differ in many fundamental ways—these are termed M and E cells for historical reasons [2]–[4] . Previous work indicates that these subgroups are functionally as well as anatomically distinct and that certain cells are associated with specific components of daily locomotor behavior . Importantly , these associations are subject to specific environmental conditions , and they display considerable plasticity under different light and temperature conditions [2] , [3] , [5]–[8] . These pacemaker subgroups communicate to synchronize with each other to produce coherent circadian rhythms [9] , [10] . Neuropeptides are critical mediators of intercellular communication between pacemaker cells in both mammals and Drosophila and a number are expressed in the Drosophila clock cell system , including the Pigment Dispersing Factor ( PDF ) [11]–[14] . Loss of the PDF peptide or its receptor leads to abnormalities in circadian locomotor behavior , including a reduction in morning anticipatory peak and a phase advance of the evening anticipatory peak under 12∶12 LD [14]–[17] . Under constant conditions these flies show high levels of arrhythmicity or short , weak rhythms . PDF controls the amplitude and phase of molecular rhythms of pacemaker cells [9] , [18] . PDF's role in synchronization of clock cells indicates that its mechanism of action is largely within the cells of the clock network . The PDF neuropeptide is expressed by two specific pacemaker subgroups ( large and small LNvs ) and the PDF receptor is expressed widely , although not uniformly , throughout the circadian network in both M and E cell groups [19] . The PDF receptor signals through calcium and cAMP , although specific signaling components remain unknown [15] , [17] . Signaling can be demonstrated in nearly all pacemaker cell groups in vivo [20] . Previous work indicates that M cells increase cAMP levels in response to at least two neuropeptides , PDF and DH31 [20] . The PDF and DH31 receptors belong to the same class II ( secretin ) G-protein coupled receptor ( GPCR ) family [21] . Both PDF [15] , [17] and DH31 receptors [22] stimulate adenylate cyclases ( AC ) to produce cAMP in vitro , and in M cells in vivo [20] , but the specific downstream components that differentiate the two peptide receptors remain unknown . Likewise , the basis for PDF's differential actions on the molecular oscillator in different pacemakers [9] , [18] has not yet been explained . The present study asks: What is the identity of downstream components that are associated with PDF-R signaling pathways in different circadian pacemaker neurons ? Specifically , using live imaging of intact fly brains , we identify the particular adenylate cyclase ( AC ) isoform that is associated with PDF signaling in small LNv—commonly called M cells . Although some signaling components are common to both DH31 and PDF neuropeptide signaling , we report that DH31 signaling does not require the same AC in the small LNv cells . This suggests that PDF signals preferentially through its favored AC , while other GPCRs , in the same identified pacemaker neurons , couple to other ACs . In addition , AC3 manipulations have no effect on PDF-R expressing LNd cells , part of the E cell network . Thus in Drosophila , critical pathways of circadian synchronization are mediated by at least two , highly specific second messenger pathways .
The Drosophila genome encodes at least 12 ACs , five of which are expressed broadly , or at least broadly in the central nervous system ( Flybase ) . The remaining cyclases ( ACXA-E and CG32301 and CG32305 ) are thought to be expressed exclusively in the male germline ( [28]; Flybase ) . Rutabaga ( Rut ) is the best characterized Drosophila ACs based on a mutagenesis screen for learning and memory phenotypes [29] . Rut is expressed in the Drosophila brain and is stimulated by calcium and calmodulin [30] . However , rut mutants showed normal PDF responses in both M and E cells ( unpublished data ) , suggesting that a different AC ( s ) must mediate PDF-dependent signaling . Nevertheless , there is evidence that cAMP signaling is involved in circadian physiology in Drosophila [31] . Therefore , to test the role of other AC isoforms in PDF responses in small LNv cells , we performed a transgenic RNAi screen using constructs directed against 11 of the 12 ACs . Initial controls were performed with and without UAS-dicer2 , however expression of dicer2 alone showed nonspecific effects on PDF responses and therefore all experiments presented were performed without dicer expression ( unpublished data ) . In M cells , two AC RNAi lines significantly reduced the amplitudes of PDF responses—AC3 and AC76E ( Figure 1C ) —although in neither case were PDF responses completely abrogated ( Figure 1C , compare to second column ) . In agreement with the initial rut mutant results , RNAi knockdown of rut mRNA had no effect on PDF responses . These results were consistent across different GAL4 lines ( Mai179-gal4 and tim ( UAS ) -gal4 ) and therefore cannot be ascribed to differences in expression pattern or strength of the specific GAL4 driver used ( unpublished data ) . The results using AC RNAi could potentially be explained by deleterious effects on small LNv exerted by continuous RNAi expression throughout the neurons' period of development . To evaluate this possibility , we employed a temperature-sensitive genetic system that allows for normal development , followed by conditional induction of RNAi only in the adult fly . Animals raised at a permissive temperature ( 18°C ) had gal4 activity blocked by a temperature-sensitive gal80 transgene ( tubulin-gal80ts ) [32] . After normal development , the flies were then moved to a higher temperature ( 29°C ) at which the gal80 transgene is no longer active and the gal4 transgene can drive expression of the RNAi construct , as well as the Epac1camps sensor . When tested in this manner , adult-specific knockdown of AC3 , but not of AC76E , resulted in a reduction of the PDF response in adult small LNv cells ( Figure 2A ) . This indicates that developmental effects likely cause the reduction observed in the initial RNAi screen for AC76C , while the reduction observed for the case of AC3 RNAi indicates its mediation of PDF responsiveness in adult small LNv pacemakers . To further confirm AC3 as the candidate PDF-dependent AC and to exclude false positives ( due to nonspecific RNAi knockdown ) , we performed further genetic tests using an independently generated AC3 RNAi line from the Harvard TRiP project ( TRiP:AC3RNAi ) in addition to the line used in the initial screen from the VDRC ( referred to as GD:AC3RNAi ) [33] that targets a non-overlapping portion of the AC3 RNA . The TRiP:AC3 RNAi line also produced a significant decrease in the amplitude of PDF responses . In addition , both the VDRC and the TRiP:AC3 RNAi lines were also tested in combination with flies that are deficient for the AC3 gene region ( Df ( 2 ) LDS6 ) , to further reduce AC3 levels . These RNAi/Df flies ( hemizygous AC3 mutants ) showed a marked further reduction of the response to PDF neuropeptide in small LNv cells compared to responses in either single mutant genotype: Df/+ or AC3 RNAi/+ ( Figure 2B ) . Together these genetic experiments provide strong confirmation of our initial RNAi screening results and support the hypothesis that AC3 is the principal mediator of PDF-dependent signaling in small LNv cells . Importantly , the consequences of knocking down AC3 were highly specific to PDF: even when combined with the deficiency , AC3 RNAi had no effect on small LNv cell responses to a closely related cAMP-generating neuropeptide—DH31 ( Figure 2C ) [20] . This indicates that , in these same neurons , DH31-R likely signals through a different AC . We tested the effects of UAS-rut , -ACXD , -AC76E , -AC3 , and -AC78C to ask whether AC over-expression could affect PDF signaling in vivo . Novel constructs were first tested for functionality by measuring cre-LUC responses to 10 µM forskolin in hEK cells . All constructs tested showed an increased average response to forskolin compared to empty vector-transfected cells , although these did not reach significance ( Figure S3 ) . We were surprised to find that , in vivo , over-expression of AC3 completely abrogated PDF responses in M cells , while over-expression of all other constructs had no such effect ( Figure 3A ) . This disruption was not due to developmental effects: delaying UAS-AC3 induction until the adult stage after completion of normal development ( using the gal80ts system ) produced the same disruption of PDF responses ( Figure S4 ) . In UAS-AC3 flies , both DH31- and dopamine-elicited cAMP increases remained intact , indicating that the cells were demonstrably healthy and could respond normally to stimulation of other Gs-coupled GPCRs ( Figure 3B and C ) . These observations suggest that abnormally high levels of AC3 specifically disrupt the PDF signaling pathway and add further proof that AC3 is a unique component of PDF signaling in M cells . Knocking down AC3 levels produced a diminution of PDF signaling in M cells ( Figure 2 ) : to evaluate further the specificity of this effect we wished to employ an AC3 rescue strategy . However , over-expressing the AC3 enzyme in M cells above normal levels disrupted M cell responsiveness to PDF ( Figure 3A ) , suggesting that supra-normal levels of the AC3 enzyme can also lead to dysfunction . Therefore , we reasoned that a successful design to rescue the AC3 knockdown would require a more moderate level of AC3 over-expression . Because the gal4 system is temperature-sensitive , intermediate levels of AC3 over-expression were achieved by raising the flies at 25°C and then moving them to 18°C overnight before imaging . This temperature shift could reduce the activity of the gal4 driver , which could result in lower levels of UAS-AC3 expression . Indeed this schedule of temperature changes reduced the disruptive effect of AC3 over-expression on responses to PDF in M cells ( Figure 4A , second column ) . We wondered whether it could also maintain effective RNAi knockdown of the endogenous gene . We confirmed that firstly the RNAi transgene is still active under this temperature regimen ( Figure 4A , third column ) . This UAS-AC3 RNAi line is directed against the 3′ untranslated region ( UTR ) of AC3 and can therefore be rescued potentially by expression of UAS-AC3 , which includes only the AC3 coding region . In fact , over-expression of UAS-AC3 , with a temperature shift from 25°C to 18°C at adulthood , rescued the reduction in PDF responses otherwise observed in a UAS-AC3 RNAi line ( Figure 4A ) . Comparable over-expression of AC78C did not rescue this deficit and that result also confirms that the rescue was not due to simple dilution of the gal4 driver . Importantly , temperature down-shifted ( 25°C to 18°C ) over-expression of AC3 alone , which should result in a small overshoot of normal enzyme levels , shows a slight reduction in PDF responses compared to control ( Figure 4A ) . This again suggested that normal levels of receptor and enzyme are key for normal function . Together these results provide strong evidence to support the hypothesis that AC3 is a specific AC isoform in M cells whose levels are tightly controlled and that normally mediates responsiveness to PDF signaling . We pursued the AC3 over-expression condition to further evaluate the nature of the components of the PDF receptor signalosome in M cell pacemakers . We reasoned that we could perhaps counteract an imbalance between signaling components produced by AC3 over-expression if we also over-expressed the PDF receptor . In fact , over-expressing PDF-R using a UAS-PDF-R transgene in combination with UAS-AC3 fully rescued the PDF response back to control levels ( Figure 4B ) . The combination of AC3 over-expression with an additional copy of PDF-R ( under control of its own promoter within a ∼70 kB transgene , termed PDF-R-myc; [19] ) produced a partial rescue of the PDF response . The latter effect was smaller than that seen with UAS-PDF-R , presumably because the induced level of PDF-R over-expression was greater with the UAS construct . Co-misexpression of the closely related neuropeptide receptor dh31-R1 ( CG17415; [22] ) along with AC3 also gave a partial rescue of diminished PDF signaling due to AC3 over-expression , although these responses were still significantly lower than control and less than what we observed with co-misexpression of PDF-R and AC3 . Together , these results suggest that ( i ) the diminution of PDF signaling that follows AC3 over-expression can be rescued by providing more PDF receptor , thus reducing the receptor/effector imbalance . It also suggests that ( ii ) the absolute ratio of PDF receptor to AC3 enzyme is important for normal neuropeptide signaling in M cells . Both RNAi and over-expression screens suggested that PDF receptor associates preferentially to the AC3 adenylate cyclase in M cells , although expression profiling studies indicate that multiple AC isoforms are expressed in these identified pacemakers [34] . To determine the specificity of AC3 contributions to other peptide signaling pathways in M cells , we evaluated cAMP responses produced by other ligands for Gsα coupled GPCRs . Drosophila DH31 ( Diuretic Hormone 31 ) is a neuropeptide closely related to mammalian Calcitonin , and the DH31 receptor ( CG17415 ) is closely related to the Calcitonin receptor [22] . Activation of PDF receptor and DH31 receptor both lead to increases in cAMP and hence both are presumably coupled to Gsα [17] , [22]; both increase cAMP in M cells in vivo [20] . RNAi knockdown of the Gsα60A subunit disrupted both signaling pathways , as expected . Interestingly , over-expression of the Drosophila G protein Gsα60A also disrupted both PDF and DH31 signaling in M cells , and responses could be restored by over-expression of the cognate receptor along with Gsα60A ( Figure 5A and B ) . As mentioned above , neither knockdown nor over-expression of AC3 affected DH31 responses ( Figure 2C ) . We interpret these results to suggest that both PDF and DH31 receptors are coupled to Gsα60A , but that PDF-R subsequently signals through AC3 and DH31-R through a different AC . Scaffolding proteins play important roles in supporting assembly of specific signalosomes , which feature tight association between specific receptors and specific second messenger molecules [35] . We hypothesized that scaffolding proteins may help explain the preference of PDF-R for coupling to AC3 . In Drosophila there are four known AKAP ( A-kinase anchoring proteins ) , molecules that bind to and help co-localize many components of cAMP signaling pathways [35] . We tested the possible involvement of AKAPs as scaffolding proteins for PDF-R in M cells using gene-specific RNAi constructs . Knockdown of the AKAP nervy , but not of the other three AKAPs , reduced PDF responses to an extent similar to that produced with the AC3 RNAi ( Figure 6A ) . As with AC3 , nervy knockdown showed no effect on DH31 responses in M cells ( Figure 6B ) . When both AC3 and nervy are knocked down together in the same M cell , PDF responses were disrupted to an even greater extent than with either RNAi alone ( Figure 6C ) . The results from single versus double RNAi constructs were generally consistent , although the comparison between TRiP:AC3RNAi and TRiP:AC3RNAi/nervyRNAi does not reach significance ( Figure 6C ) . This suggests that nervy also plays a role in PDF signaling in small M cells , presumably by allowing PDF signaling components to effectively localize and thus promote efficient signaling . A number of previous studies have suggested that PDF signaling pathways differ between M and E cells . We therefore investigated PDF-R expressing LNd cell ( the CRY+/PDF-R+ subset of LNd , using the Mai179-gal4 driver [18] , [19] , [36] ) to evaluate the role of AC3 in E cell PDF signaling . We first confirmed that PDF-induced cAMP responses in these neurons are dependent upon PDF-R; flies with the strong PDF-R mutation ( han5304 ) totally lose E cell responsiveness ( as has been previously reported in M cells ) [20] . As in M cells , Gsα manipulations again reduced PDF responses in E cells ( Figure 7B ) . Previous experiments confirmed that ACs are involved in E cell PDF responses ( Figure S2 ) . The first AC we tested was rutabaga , which had proven ineffective in reducing PDF responses in M cells ( Figure 1C ) . E cells in rut mutants produced normal PDF responses ( unpublished data ) ; responses to PDF were likewise normal following rut RNAi expression ( Figure 7C ) . In the case of AC3 , neither RNAi knockdown ( combined with a AC3 Df ) nor AC3 over-expression altered PDF responses in this E-type clock cell subgroup . E cell responses were robust even though these same genetic manipulations produced the most severe reductions in M cell PDF responses ( compare Figure 7C with Figure 2B and Figure 3A ) . Notably , M cell responses were reduced even when measured in the same brains in which E cells proved responsive ( unpublished data ) . The foregoing data argue that AC3 mediates the cAMP generation produced by PDF in M cells . To what extent is circadian locomotor behavior affected by this disruption of this AC3 activity ? Manipulations that partially reduced M cell responses to PDF ( e . g . , RNAi knockdown of any single AC or AKAP ) did not affect locomotor rhythms ( see Figure S5 and Tables 1 and 2 ) . However , combining AC3 RNAi knockdown with a deficiency for the AC3 region produced a very strong reduction in the morning anticipation peak , as well as higher levels of arrhythmicity under constant conditions ( Figure 8B and Tables 1 and 2 ) . We observed the same effects in UAS-AC3 over-expression in PDF cells ( Figure 8C and Tables 1 and 2 ) . Over-expression of UAS-PDF-R and UAS-AC3 together slightly reduced arrythmicity in DD compared to UAS-AC3 alone ( Table 2 ) . However , the loss of morning anticipation seen in the UAS-AC3 condition is not rescued by over-expression of the PDF receptor ( Figure 8D and Table 1 ) . This suggests that , although the PDF FRET response is rescued ( Figure 4B ) , additional ( for example , temporal ) aspects of PDF signaling may contribute to normal circadian behavior in LD .
Networks of pacemaker cells are synchronized by intercellular interactions [3] , [9] , [19] . There is strong and diverse evidence that control of cAMP levels is a critical factor underlying pacemaker rhythmicity and synchronization . Daily changes in cAMP levels in SCN neurons contribute to setting the phase , period , and amplitude of PER2 cycles and thus represent an integral component of the clock mechanism itself [37] . Furthermore , the RGS16 regulator sets the level of cAMP generation and its levels are likewise clock-controlled [38] . Regarding synchronizing agents that couple diverse pacemakers , both PDF in the fly and VIP in the mouse produce cAMP increases in response to receptor activation , and these signals ultimately have access to the pacemaker mechanism in target cells [4] , [9] , [10] , [39]–[41] . Thus understanding the molecular components that control cAMP metabolism in pacemaker neurons , especially those downstream of receptors for the PDF and VIP modulators , are significant goals for the field . There are at least 12 different genes encoding adenylate cyclases in the fly genome , of which the best known is Rutabaga , a calcium- and calmodulin-sensitive AC . Rut was first identified in a screen for mutations that affected learning and memory exhibited in an associative conditioning paradigm [29] . The Rut cyclase displays the properties of a coincidence detector with its activity triggered by inputs from simultaneous activation of more than one GPCR [24] . However , our studies indicate that , in M pacemakers , the PDF receptor is preferentially coupled not to Rut but to the adenylate cyclase encoded by AC3 . In vitro studies suggest the AC3 cyclase may be inhibited by calcium [42] . The functional consequences of this specific signaling association , the physical basis that supports it , and the degree to which it may hold true in other PDF-responsive neurons in the Drosophila brain are important questions raised by this work . The experiments that manipulated AC and PDF-R expression together indicate that relative levels of AC enzyme and receptor are important determinants of normal PDF cAMP responses in M pacemakers . Counterintuitively , AC3 over-expression was as effective in diminishing PDF responsiveness as was AC3 knockdown . One possible explanation is that the abnormally high levels of AC3 result in incorrect subcellular localization of signaling components , which may preclude the ability of AC3 to contribute to cAMP generation . Within M cells , only moderate expression of a UAS-AC3 transgene could restore normal PDF responses after knockdown of endogenous AC3 . Likewise , over-expressing AC3 together with PDF-R could restore the balance between receptor and effector , as indicated by the return of PDF responsiveness . Although these results may not generalize to all cell types or receptor pathways , it is notable that , for this circadian signaling pathway , appropriate levels of signaling components were as important as their simple presence or absence . The reliance on proper stoichiometry between receptor and AC is further evidence to support the hypothesis that PDF-R and AC3 exhibit a specific functional association within the M class of pacemaker cells . One possible explanation for preferential coupling of PDF-R to AC3 is simply that it is the only adenylate cyclases to be expressed in M cells . However this explanation is inconsistent with at least two notable observations—first , M cells in flies with a severe AC3 knockdown ( Df2L;GD:AC3RNAi ) still elevate cAMP levels normally in response to neuropeptide DH31 . Second , according to recent profiling studies , multiple other ACs are normally expressed at appreciable levels in larval LNs [43] and in adult LNv [34] . Interestingly , these studies indicate that AC3 is not even the most abundant adenylate cyclase [34] . Therefore , we favor an alternative explanation—that molecular specificity dictates the composition of different receptor pathways , with PDF-R residing in privileged association with AC3 ( Figure 9 ) . There is clear support for the concept of preferential coupling between GPCRs and specific ACs in multiple cell types , in addition to our own findings in Drosophila clock cells . Previous work in Drosophila [44] suggests that individual cyclases play specific roles in G-protein signaling associated with gustation . Furthermore , studies of the GABAergic system in the mouse pituitary indicate that Type 7 adenylate cyclase is associated with ethanol and CRF sensitivity , although mRNA for four of the nine mammalian ACs are detected by microarray in pituitary tissue [45] , [46] . It has also been proposed that receptor/AC preference may depend upon environmental conditions: for example , the Type 7 preference of the CRF receptor in the mouse amygdala occurs only after phosphorylation of signaling components . Without phosphorylation , CRF receptor couples preferentially to Type 9 adenylate cyclase [47] . Thus , our results add to the body of evidence that highly specific associations between receptors and their downstream partners are key regulators of signaling . There is clear evidence that signaling components within specific pathways do cluster , which may explain how generalized signaling molecules like cAMP and PKA are capable of targeting distinct downstream effectors . Much current work focuses on possible mechanisms for such localization , [35] and the concept of signalosomes has been proposed to describe the spatial sequestering of signaling pathway components to promote exactly this sort of specific association . Thus preferential AC3/PDF-R coupling may be achieved by localizing AC3 near to PDF receptors . Mechanisms for grouping signaling components may include their co-localization in lipid rafts; many of the components of cAMP signaling including G proteins , PDE , PKA , and cyclic nucleotide gated channels are found in lipid rafts [48] and studies in human bronchial smooth muscle cells detected three different AC isoforms , which are present in distinct membrane microdomains and which respond to different neurotransmitters and hormones [49] . In addition , it is likely that another clustering mechanism includes the formation of macromolecular structures through the use of scaffolding proteins that bind to signaling molecules , as first proposed by Stadel and Crooke [50] . Later studies showed that ACs form large complexes with β-arrestins , G proteins , and calcium channels [51] . The scaffolding protein InaD is required for normal localization of signaling components in the fly visual system including TRP and PLC [52]–[54] . Specialized signaling components such as AKAPs ( A-kinase anchoring proteins ) can bind to receptors as well as kinases and adenylate cyclases [35] . In Drosophila , AKAPs organize functionally discrete pools of PKA , and disruption of these signaling complexes alters normal spatio-temporal signal integration and causes a loss of anesthesia-sensitive as well as long-term olfactory memory formation in flies [55] , [56] . In our study , knockdown of AKAP nervy reduced PDF responses: These results lead to a hypothesis whereby , in M pacemakers , PDF receptor preferentially couples to AC3 via a nervy-based scaffold system to produce normal circadian behavior ( Figure 9 ) . We emphasize that , while our results demonstrate a functional connection between AC3 and PDF-R , the basis for any physical connections has not yet been established . Although our study provides an example of a specific receptor/enzyme pairing in a subset of circadian clock cells , our evidence also suggests the exact details of PDF signaling in other Drosophila pacemakers may differ . Simply put , the set of AC3 manipulations that caused a disruption of PDF responsiveness in M pacemakers had no such effect in E pacemakers . Importantly , disruption of Gsα affected both subgroups ( see Figures 5A and 7B ) . Multiple lines of evidence have suggested that PDF signaling differs between clock cell subgroups . ( i ) Loss of PDF has distinct effects on PERIOD protein cycling in LNv ( M cells ) versus non-LNv cells ( E cells ) . Both cell groups continued to show cycling in PER immunostaining levels and localization but , while M cells become phase-dispersed in PER cycles , E cells remain synchronized with altered phase and amplitude of PER accumulation [9] , [16] . ( ii ) In Pdf/cry and PDF-R/cry double mutants , a subset of E cells show a severe attenuation of the PER molecular rhythm , while M cells continue to cycle normally [41] , [57] , [58] . Different subsets of E cells have previously been implicated in control of evening anticipation , and even when AC3 is altered in all clock cells , the evening peak retains its proper phase , again suggesting that AC3 is not a required enzyme in E type pacemaker cells ( unpublished ) . These finding are consistent with the hypothesis that there are two functionally different PDF signaling pathways . However , although we have confirmed that adenylate cyclases are responsible for the PDF FRET responses in E cells ( Figure S2 ) , as yet we have no positive evidence regarding the contribution of any single AC in E pacemakers ( unpublished data ) . Hence it remains to be determined how uniform the components of PDF signalosomes in the M versus E pacemaker cell types are . How well do the observations obtained with neuronal imaging predict or correlate with circadian locomotor behavior ? Manipulations of AC3 that severely disrupt PDF signaling in M cells were correlated with a loss of morning anticipation and increased arrythmicity in DD . Manipulations that only partially reduce the FRET response ( for example , single AC3 or single nervy knockdown ) resulted in normal circadian locomotor behavior or disruptions to some aspects but not all . The latter observations suggest that the animal is capable of compensating for reduced AC3-generated cAMP responses by M cells but not to complete loss of AC3 function ( see Tables 1 and 2 and Figure S5 ) . These data argue for a contribution to behavior by PDF signaling via AC3 in M cells and stand in contrast to a recent report by Lear et al . [10] . That group reported that PDF-R expression in E cells alone is sufficient for morning anticipation and that exclusive expression of PDF-R in M cells does not recover morning anticipation . We cannot reconcile these differences without further experimental efforts , but note that GAL80 techniques are not always sufficient to extinguish gene expression in vivo ( unpublished data ) . Depending on ambient conditions [12] , [57] , the M cells contribute to normal morning anticipatory behavior and to maintenance of rhythmicity under constant dark conditions [2] , [14] , [20] , [58]–[60] . However , in our study M cells expressing AC RNAi remain normally responsive to at least two other neurotransmitters ( DH31 and dopamine ) . Hence we suspect that much of the behavioral effect of knocking down AC3 in M pacemakers is mainly due to loss of PDF signaling in them despite retention of additional inputs from a PDF-independent source . Levels of PDF receptor and responsiveness to PDF are both high in small LNv cells and absent ( or barely detectable ) in large LNv cells [19] , [20] , [34] . Therefore we expect that AC3 alterations in M cells ( directed by Pdf-GAL4 ) primarily affect PDF signaling in LNvs . In these considerations , the extent to which the AC3 behavioral phenotype is explained by PDF-R coupling to AC3 in M cells is not yet defined . AC3 appears coupled to at least one other GPCR pathway in LNvs because , in DD , AC3 knockdowns produced a more severe behavioral phenotype than did Pdf null flies ( a higher percentage of arrhythmicity ) . Knockdown of Gsα60A levels of the M pacemakers lengthened the period in DD , a behavioral effect opposite to those seen following loss of PDF , or M cell ablation , namely . Previous studies of Gsα60A in M cells also reported a long period phenotype [43] . Likewise selective expression in small LNv of shibiri ( a dominant negative allele of the fly homolog to dynamin [61] ) or of a chronically open sodium channel [60] both produce long period phenotypes [62] , [63] . Although we cannot rule out a PDF-dependent role in period lengthening in our Gsα60A experiments , our imaging data suggest the lengthened period phenotype may be explained by the fact that alterations of Gsα60A impact multiple signaling pathways ( see Figure 5 ) . Our results demonstrate in Drosophila that , in small LNv ( M ) circadian pacemakers , a highly specific signaling cascade is activated in response to PDF . They suggest there exists a dedicated PDF-R::AC3-dependent signaling pathway that functions to synchronize these particular clock cells . A different PDF signaling cascade is likely to operate in E pacemakers . The complete molecular details of these signaling complexes , their convergence with CRY signaling [41] , and their ultimate connections to the cycling mechanism are significant issues for future studies .
Drosophila were reared on cornmeal/agar supplemented with yeast and reared at 25°C , unless otherwise indicated by experimental design . Male flies ( age 2 to 5 d old ) were moved to 29°C for 24–48 h before imaging to increase UAS transgene expression . For temperature shift ( tubulin-gal80ts ) experiments , crosses were maintained at 18°C to maintain gal80ts suppression of gal4 , and males were collected and moved to 29°C for 24–48 h before imaging to allow UAS transgene expression . For temperature shift UASAC3/TRiP:AC3RNAi rescue experiments , males were reared at 25°C and moved to 18°C for 12–16 h before imaging to reduce gal4-driven expression of AC3 . All gal4 lines used in this study have been described previously: Pdf ( m ) -gal4 [64] , UAS- Epac1camps50A [20] , and Mai179-gal4 [65] . The TRiP:RNAi ( UAS-TRiP:AC3RNAi , UAS-TRiP:-nervyRNAi , UAS-TRiP:AKAP200RNAi ) , UAS Gsα60A , UAS-rutabaga , tubulin-gal80ts , and Df ( 2 ) LDS6 lines were obtained through the Bloomington Stock Center ( thanks to the Harvard TRiP RNAi project ) and the UAS-Gsα60ARNAi , UAS-GD:AC3RNAi , UAS-AC13ERNAi , UAS- AC78C , UAS-rutRNAi , UAS-ACXARNAi , UASACXBRNAi , UAS-ACXCRNAi , and UASACXDRNAi . UAS-yuRNAi and UAS-rugoseRNAi lines were obtained through the Vienna RNAi Stock Center . For epifluorescent FRET imaging , living brains expressing gal4-driven uas-Epac1camps were dissected under ice-cold calcium-free fly saline ( 46 mM NaCl , 5 mM KCl , and 10 mM Tris ( pH 7 . 2 ) ) . All lines tested included one copy each of gal4 ( Pdf-gal4 used for small LNv cells and Mai179gal4 for PDF-R ( + ) LNd cells ) and Epac1camps . All genotypes include one copy of each transgene unless otherwise indicated . Full genotypes are available in Table S1 . For the RNAi AC screen and for pharmacological experiments , whole brains were placed at the bottom of a 35×10 mm plastic FALCON Petri dish ( Becton Dickenson Labware ) as in [20] , incubated in HL3 saline , and substances tested by bath application . For all remaining experiments , dissected brains were placed on poly-l-lysine-coated coverslips in an imaging chamber ( Warner Instruments ) , and HL3 was perfused over the preparation ( 0 . 5 mL/minute ) . Microscopy was performed through a LUMPL 60×/1 . 10 water objective with immersion cone and correction collar ( Olympus ) on a Zeiss Axioplan microscope . Excitation and emission filter wheels were driven by a Lambda 10-3 optical filter changer and shutter control system ( Sutter Instrument Company ) and controlled with SLIDEBOOK 4 . 1 software ( Intelligent Imaging Innovations ) . Images were captured on a Hamamatsu Orca ER cooled CCD camera ( Hamamatsu Photonics ) . Exposure times were 20 ms for YFP- FRET and 500 ms for CFP donor . Live FRET imaging was performed on individual cell bodies , YFP-FRET and CFP donor images were captured every 5 s with YFP , and CFP images were captured sequentially at each time point . Following 45 s of baseline YFP/CFP measurement the peptide was bath added/injected into the perfusion line to result in a final concentration of 10−06 M . FRET readings were then continued to result in a total imaging time course of 10 min . ODQ and dopamine were purchased from Sigma . Synthetic DH31was provided by David Schooley and PDF was produced by ( Neo MPS , San Diego , CA , USA ) . For all experiments reported , we collected responses from at least 10 cells that were found in at least five brains for all genotypes . A region of interest ( ROI ) defined each individual neuron , and for each , we recorded background-subtracted CFP and YFP intensities . The ratio of YFP/CFP emission was determined after subtracting CFP spillover into the YFP channel from the YFP intensity as in [26] . The CFP spillover ( SO ) into the YFP channel was measured as . 397 [20] . For each time point , FRET was calculated as ( YFP− ( CFP * SO CFP ) ) /CFP . To compare FRET time courses across different experiments , FRET levels were normalized to initial baseline levels and smoothed using a 7-point boxcar moving average over the 10-min imaging time course . Statistical analysis was performed at maximal deflection from the initial time point by performing ANOVA analysis followed by post hoc Tukey tests using Prism 5 . 0 ( Graphpad Software Inc . ) . Over-expression constructs were built by PCR construction from cDNA derived from adult heads ( Canton S ) and subcloned into P{cDNA3} and P{UAS-attb} vectors . The original AC3 clone was a kind gift from Lonny Levin ( Weill Cornell Medical College ) . The sequences of all primers used in this study are: AC3 ( BamHI ) 5′: GGATCCATGGAAGCAAATTTGGAGAACGGTC; AC3 ( EcoRV ) 3′: GATATCCTATTCTAGCAAAGACTGACATTCT; AC78C 3′: CTATAACGCATCGTTGTGGCTCTTCGATAT; AC78C nested 3′: ACTTAGACCCAGTGAGTGCGCGTACTCGG ; AC78C 5′: ATGGACGTGGAACTCGAAGAGGAGGAGGAG; AC78C nested 5′: GCATAGCAATAGACAGAATCCTCCGCCACA; AC76E 3′: CTACAATTTCCCATCGAAAGGTGTCTTTAC; AC76E nested 3′: ATCAACAGCAACTGGGTGACGATCGGTGAT; AC76E 5′: ATGGTAAATCACAATGCGGAAACTGCGAAA; AC76E nested 5′: GCCACTAGCTACACGCCACCGCTTTTCGCC; ACXD5′: ATGGACTCCTACTTCGACTCGGCC; and ACXD3′: CTAGTCTTCTTTGGTTGGCGCGGCC . hEK-293 cells were tested using a cre-LUC reporter in response to 10 µM forskolin 24 h post-transfection with different UAS-AC constructs that had been subcloned into p{CDNA3} . All constructs were co-transfected with cre-luc and compared to empty- vector-transfected cells ( 0 . 5 µg cre-luc and 2 . 5 µg PDF-R and 2 . 5 µg AC ) . Four hours after forskolin addition , cells were lysed and luciferin added , followed by bioluminescence measurement using a Victor-Wallac plate reader . Measurements were performed in triplicate and normalized to vehicle-treated controls; the results represent combined activities from three independent transfections . Male flies were loaded into Trikinetics Activity Monitors 4–6 d after eclosion . Locomotor activities were monitored for 6 d under 12∶12 light/dark and then for 9 d under constant darkness ( DD ) conditions . Anticipation index was calculated as in [19] as ( activity for 3 h before lights-on ) / ( activity for 6 h before lights-on ) . To analyze rhythmicity under constant conditions we normalized activity from DD Days 3–9 and used X2 periodogram with a 95% confidence cutoff as well as SNR analysis [66] . Arrhythmic flies were defined by having a power value <10 . | In the fruit fly Drosophila melanogaster , the neuropeptide Pigment Dispersing Factor ( PDF ) supports circadian function by synchronizing two types of pacemaker cells , M cells and E cells . The PDF receptor ( PDF-R ) is a G protein coupled receptor ( GPCR ) whose activation stimulates adenylate cyclase ( AC ) , thereby elevating levels of the second messenger cAMP in many different pacemakers including M cells . Drosophila contains at least 12 genes that encode potential ACs . In this study , we identify the AC downstream of the PDF receptor specifically in M cells and show that PDF signals preferentially through AC3 . However , other GPCRs in the very same cells do not rely on AC3 . A different scaffolding protein also influences PDF responses in M cells , suggesting that signaling components are spatially grouped to allow for coupling of specific receptors with downstream components . Remarkably , in E pacemakers , AC3 disruptions have no effect . These findings suggest that distinct PDF circadian signals exist in M versus in E pacemakers , and more generally , we propose a mechanism to differentiate signaling pathways that use common components . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"molecular",
"neuroscience",
"adenylyl",
"cyclase",
"signaling",
"pathway",
"second",
"messenger",
"system",
"mechanisms",
"of",
"signal",
"transduction",
"neuroscience",
"neurotransmitter",
"receptor",
"signaling",
"membrane",
"receptor",
"signaling",
"signaling",
"pathways",
"neurotransmitters",
"biology",
"molecular",
"biology",
"signal",
"transduction",
"cellular",
"neuroscience",
"molecular",
"cell",
"biology",
"behavioral",
"neuroscience"
] | 2012 | The Circadian Neuropeptide PDF Signals Preferentially through a Specific Adenylate Cyclase Isoform AC3 in M Pacemakers of Drosophila |
Recent high-throughput efforts such as ENCODE have generated a large body of genome-scale transcriptional data in multiple conditions ( e . g . , cell-types and disease states ) . Leveraging these data is especially important for network-based approaches to human disease , for instance to identify coherent transcriptional modules ( subnetworks ) that can inform functional disease mechanisms and pathological pathways . Yet , genome-scale network analysis across conditions is significantly hampered by the paucity of robust and computationally-efficient methods . Building on the Higher-Order Generalized Singular Value Decomposition , we introduce a new algorithmic approach for efficient , parameter-free and reproducible identification of network-modules simultaneously across multiple conditions . Our method can accommodate weighted ( and unweighted ) networks of any size and can similarly use co-expression or raw gene expression input data , without hinging upon the definition and stability of the correlation used to assess gene co-expression . In simulation studies , we demonstrated distinctive advantages of our method over existing methods , which was able to recover accurately both common and condition-specific network-modules without entailing ad-hoc input parameters as required by other approaches . We applied our method to genome-scale and multi-tissue transcriptomic datasets from rats ( microarray-based ) and humans ( mRNA-sequencing-based ) and identified several common and tissue-specific subnetworks with functional significance , which were not detected by other methods . In humans we recapitulated the crosstalk between cell-cycle progression and cell-extracellular matrix interactions processes in ventricular zones during neocortex expansion and further , we uncovered pathways related to development of later cognitive functions in the cortical plate of the developing brain which were previously unappreciated . Analyses of seven rat tissues identified a multi-tissue subnetwork of co-expressed heat shock protein ( Hsp ) and cardiomyopathy genes ( Bag3 , Cryab , Kras , Emd , Plec ) , which was significantly replicated using separate failing heart and liver gene expression datasets in humans , thus revealing a conserved functional role for Hsp genes in cardiovascular disease .
The increasingly cheaper and rapid accumulation of large -omics datasets across several experimental conditions has prompted generation of a wealth of data on biological networks . This growth of network data now permits their large scale applications to biomedical research , including analysis of gene function , metabolic and signaling pathways , as well as disease-related or cell function-related networks [1] , [2] . However , reconstructing and interpreting large biological networks , such as co-expression networks , protein-protein interaction networks or genetic networks , with different features ( e . g . , sparse or densely interconnected , etc . ) poses many challenges , advocating efficient and flexible methods for network inference and pattern discovery . An important level of complexity in current network analysis regards its extension to multiple conditions , for instance different species [3] , cell-types [4] or disease states [5] , [6] . For example , reconstruction of networks across multiple disease-states is becoming a useful approach for efficient drug-target discovery , as networks can inform the “biological context” ( e . g . , pathways , cellular processes ) where genes operate and therefore can help designing better therapeutic interventions [7] . In genetic studies of complex diseases researchers increasingly focus on groups of highly interconnected genes within larger networks ( referred to as clusters , modules or subnetworks ) to elucidate specific cellular and molecular processes that might represent functional disease mechanisms and pathological pathways [8]–[10] . While several computational tools for network analysis in single datasets or conditions are available , only few computationally efficient methods for genome-scale network analysis across multiple conditions have been developed to date . These methods can be broadly classified into two main categories: ( i ) methods to find the “difference” between networks across conditions or to pinpoint condition-specific networks [11]–[14] , or ( ii ) methods to identify the common parts in networks across conditions [15]–[17] . More recently , tensor-based computational frameworks [15] or probabilistic Markov blanket search algorithms [18] have been proposed to learn network structures across conditions . However , these methods are either heavily influenced by the choice of input parameters ( e . g . , number of clusters , number of nodes within a cluster , cluster interconnectivity ) [15] or , being based on probabilistic graphical modelling , they become prohibitively slow for high number of conditions since they are trying to learn the structure of large graphs [18] . Complementary to the above approaches , spectral methods , such as Singular Value Decomposition ( SVD ) , have been also proposed to investigate patterns of connectivity between nodes within a single network [19] , [20] or for comparing two networks [21] . Generally , any network can be described as a graph , which is denoted as comprising a set of vertices or nodes together with a set of edges [22] . The graph may be represented by a square , symmetric , real-valued matrix of size whose entries denote the relationship between the corresponding nodes . In the affinity matrix , the element , called weight , represents the strength of connection between vertices and . For instance , in gene regulatory ( or co-expression ) networks , the nodes might represent genes ( or mRNAs expression ) and edges represent the strength of gene-gene interactions ( or mRNAs co-expression ) . Generalized Singular Value Decomposition ( GSVD ) can be used to identify sub-network structures and for comparative analysis of genomic datasets across two conditions [11] , [23] . Given two matrices and [24] , [25] , their GSVD is given by ( 1 ) where and have orthonormal columns , is invertible , with , with . The ratios are the generalized singular values of and . In this setup , the common factor is informative of the cluster structure shared across the two data matrices . Recently , a novel mathematical formulation , higher-order GSVD ( HO GSVD ) , which is constructed for more than two data matrices has been proposed [26] . Under this framework , the matrices , each with full column rank ( i . e . , the maximum number of linearly independent column vectors of is ) , are decomposed as ( 2 ) where is composed of normalized left basis vectors , with and the latent factor matrix is composed of normalized right basis vectors . The HO GSVD can be also derived in the special case of square , symmetric , full rank affinity matrices , , where each element represents the weight of the edge between node and in the th condition . It has been previously employed to compare multiple datasets with identical column size in order to detect their common substructures of columns ( i . e . , observations ) [26] . Yet , another useful application of the HO GSVD to genomics is to set it to discover gene networks across multiple conditions and pinpoint “common” and “differential” cluster structures . In this paper , we build on the flexible HO GSVD mathematical framework and propose a new , parameter-free computational algorithm ( Cross-Conditions Cluster Detection or C3D ) for automatic detection of both similarity and dissimilarity clustering patterns in large weighted ( and unweighted ) networks across several conditions ( ) . The original HO GSVD model has been employed for analysis of datasets that had varying number of genes ( ) , the same number of observations ( ) ( i . e . , arrays/time points in [26] ) across conditions and with . As such , this illustrative application of the HO GSVD in genomics was aimed at the identification of common structures within the observations [26] . Here , we built on the initial HO GSVD to extract sub-structures ( i . e . , common and differential clusters ) from genes across multiple conditions ( ) by applying the decomposition to the transposed expression matrix . We show how this enables a more general application of the HO GSVD framework to genome-scale network analysis of genomic data ( e . g . , microarray , RNA-seq ) in multiple conditions . Besides , a distinctive feature of our method is in its capability to take as an input either the raw expression matrices or co-expression matrices , allowing flexibility in the choice of the co-expression measures ( e . g . , Spearman , Kendall , mutual information , etc . ) . Figure 1 illustrates the working principle of the C3D algorithm . The input data for C3D can be provided into different formats to be used by the HO GSVD: ( i ) the raw expression data matrices ( ) or ( ii ) the co-expression data matrices ( ) . In the former case , a first data initialization step is conducted where the input expression matrices , with the same number of genes are converted to co-expression matrices by scaling their variance to 1 and taking their quadratic form . In the second step ( HO GSVD-based algorithm ) , an approximate HO GSVD is employed to identify a common basis , with representing the dimension of the GSVD common subspace , for the decomposition of the input datasets and identify the common and differential correlation structures . The HO GSVD-based algorithm computes a square matrix , which is built on the arithmetic mean of all pairwise quotients where denotes the Moore-Penrose inverse of the co-expression matrix [24] ( see Methods section ) . The first eigenvectors of ( according to the norm of the corresponding eigenvalues ) are then used to identify an approximate decomposition of the input co-expression matrices and form the decomposition basis . Specifically , each selected column vector of is used to reorder the input data matrices such that candidate “common” ( or “differential” ) clusters can be identified . In the third step ( cluster nodes selection and validation ) , we employ a mixture model approach to classify genes and assign them to each cluster based on a misclassification error rate ( MER ) . Finally , we implemented an empirical cluster validation procedure to identify the conditions where clusters are present and assess the level of significance for clusters within each condition . To demonstrate the increased power and benefits of our HO GSVD-based algorithm , we carried out an extensive simulation study and benchmarked C3D against commonly used methods that were designed to detect either common ( WGCNA [16] , [17] ) or differential network structures ( DiffCoEx [13] ) across multiple conditions . We show that our approach has higher power and stability in detecting both common and differential co-expression clusters across all simulated conditions , while being two to seven fold less computationally intensive than alternative methods . In contrast with alternative approaches that require specification of ad-hoc input parameters , the proposed method has the distinctive advantage of being parameter-free , which makes it a powerful tool for real data exploration and analysis . To substantiate this claim , we applied C3D to publicly available transcriptomic datasets in rats and humans and identified several multi-tissue gene co-expression networks that were associated with specific functional processes relevant to phenotypic variation and disease .
We carried out a simulation study to compare our method with commonly used approaches for identification of “common” or “differential” clusters across multiple networks: ( 1 ) WGCNA and ( 2 ) DiffCoEx . The WGCNA method for detection of common clusters across co-expression networks employs a “soft” threshold to assign a connection weight to each gene pair and extract densely connected gene clusters that are present in all conditions . The DiffCoEx method follows a strategy similar to WGCNA but , instead , it focuses on detecting the differences in co-expression patterns ( “differential” clusters ) between multiple conditions . Additional details on the specific parameterizations used in for WGCNA and DiffCoEx analyzes are reported in Text S1 . To simulate a realistic example of gene expression data from multiple conditions that represent a typical “small large ” scenario , we draw inspiration from a publicly available multi-tissue microarray dataset consisting of genome-wide expression profiles from recombinant inbred rat strains in seven tissues [27] . We simulated different types of clusters that are either detected in all conditions ( “common” clusters ) or are specific to a subset of conditions ( “differential” clusters ) , Figure 2 . We considered dense clusters of variable sizes ( 100–500 nodes ) where each node is connected with all other nodes in the cluster with a given weight ( ) , which is defined as the Pearson correlation between expression profiles of genes and . We simulated clusters with varying cluster densities ( 0 . 1 , 0 . 3 , 0 . 5 , 0 . 7 ) , which were defined as the average Pearson correlation between any pair of nodes within a cluster . In addition to the simple case of a cluster common to all conditions and with the same size ( Cluster pattern 1 ) , we set out to evaluate the sensitivity of our and alternative approaches to detect clusters which are present only in a subset of conditions and that overlap partially across conditions . This is more likely to be relevant for analysis of pathways and gene networks across tissues or during development , where varying gene-sets can exert their function only at specific developmental times or in specific cell-types . To account for these more complex scenarios , we simulated “nested” ( Cluster pattern 2 ) and partially “overlapping” ( Cluster pattern 3 ) cluster structures ( Figure 2 ) . Cluster pattern 2 and Cluster pattern 3 have an intersection part , defined by the nodes in common to all conditions , and a union part , defined by the nodes in common to all conditions plus the nodes present in individual conditions . In summary , for each of the four cluster densities considered one dataset consisted of a and matrix in conditions , where each cluster type ( Clusters patterns 1–3 ) was simultaneously present in the data matrix . To assess reliability of the results , for each of these data we generated 20 independent replicates , yielding a total of 560 simulated datasets . Similarly , to evaluate how the number of available observations affects the methods' performance we simulated datasets consisting of a and matrix in conditions ( 20 replicates , 560 datasets in total ) . See Text S1 for additional details . The True Positive Rate ( TPR ) and the False Positive Rate ( FPR ) are widely used as evaluation metrics for a classification model and can be used to quantitatively assess ( and compare ) methods performance [28] . The TPR defines how many correct positive results ( simulated clusters genes within the called cluster ) occur among all results called positive in the analysis by a given method . FPR , on the other hand , defines how many incorrect positive results occur among all results called positives . Typically , a and the corresponding indicate a perfect classifier ( or a perfect method ) . In our simulation study , the best cluster detection method would yield both high TPR and low FPR levels for different cluster types , sizes and densities . For each simulated cluster type , Figure 3 shows the TP/FP rates for C3D , WGCNA and DiffCoEx methods as a function of the simulated cluster densities . For C3D we controlled the ( local ) misclassification error ( i . e . , the probability to assign wrongly a gene to a cluster ) to be less than 0 . 05 or less than 0 . 2 , and required that each cluster is detected with , whereas for WGCNA and DiffCoEx we used two ( default ) parameterizations chosen according to the software guidelines ( see Methods section ) . The C3D method outperformed WGCNA in the identification of clusters present in all conditions ( Cluster pattern 1 , Figure 3 ) , and showed to have consistently high TPR ( and very low FPR , ) irrespective of the simulated cluster density . WGCNA performance varied considerably as a function of the simulated cluster density and , depending on the adopted parameterization , FPR levels were ( reaching 20% in one case ) , Figure 3 . Furthermore , we observed large variations in WGCNA performance ( mostly in the TPR ) , which are indicated by the large standard deviations in TPRs calculated from the 20 replicated datasets . For more complicated patterns ( “nested” and “overlapping” clusters ) , we compared C3D with WGCNA to detect the intersection part ( 100 nodes ) of common clusters . Since WGCNA is designed to detect only those clusters shared across all conditions , for clusters present in a subset of conditions , we run WGCNA only in the set of conditions where the simulated clusters were present . For Cluster patterns 2–3 , C3D and WGCNA performances were similar , reaching high TPR for detection of the intersection part of clusters with simulated ( Figure 3 ) . However , C3D showed higher TPRs than WGCNA to detect clusters with low densities ( 0 . 1–0 . 3 ) , while controlling the FPR at low levels ( , Cluster pattern 2 intersection ) . In the case of partially overlapping clusters present in a subset of conditions ( Cluster patterns 2–3 ) we compared C3D with DiffCoEx in respect of detecting the union part ( 500 nodes ) of “differential” clusters , and calculated TPR and FPR for detection of this cluster ( indicated with a black square at the top of Figure 3 ) . We found that C3D outperformed DiffCoEx across the simulated scenarios . In the case of the “nested” cluster structures that are present in 5 out of 7 conditions , C3D had consistently higher TPR levels than DiffCoEx , which showed comparable TPR levels only for detection of highly-dense clusters ( i . e . , , Cluster pattern 2 union , Figure 3 ) . However , similarly to what observed for WGCNA method , in this case DiffCoEx showed large variability in its performance across the 20 replicated datasets . The difference in performance between C3D and DiffCoEx was observed also in the more complicated case of partially overlapping cluster structures ( Cluster pattern 3 ) . In this case , C3D showed consistently higher TPR than DiffCoEx that reached a maximum as compared with of C3D . Both methods showed comparably low FPR ( ) for detection of the union part of Cluster patterns 2–3 ( Figure 3 ) . Similarly to what observed for the simulated data with observations , C3D performed better than ( or as good as ) both WGCNA and DiffCoEx when benchmarked on simulated datasets with only observations ( Figure S1 ) . As expected , all methods had lower TPRs associated with the detection of low-density clusters , however also with a small number of observations , C3D showed significantly better ( and more stable ) results than WGCNA and similar performance as compared with DiffCoEx . Notably , for detection of “common” clusters present in all conditions ( Cluster pattern 1 ) , CD3 held high TPR levels ( and FPR ) whereas WGCNA's performance dropped significantly , reaching a maximum TPR ( Figure S1 ) . These data show that C3D on balance performed better than WGCNA and DiffCoEx across all simulated scenarios . We underline that while WGCNA and DiffCoEx methods are specifically designed to detect either common or differential clusters , respectively , here we showed that C3D was equally or more accurate than both methods in the detection of common and differential cluster structures . We also highlight how C3D ability to detect correctly the simulated clusters was highly consistent across all runs on the replicated datasets , as shown by the small standard deviations of the mean TP and FP estimates ( Figure 3 ) . In contrast , we observed that both WGCNA and DiffCoEx performances varied appreciably across the replicated simulations , often resulting in large standard deviations of the mean TP and FP estimates . To better assess the reliability of the different methods we calculated the relative standard deviation of the TPR measured in all analyzed datasets . In 560 simulated datasets of size , the C3D method had a median RSD of ( range 113 . 36 ) whereas WGCNA and DiffCoEx have median ( range 447 . 2 ) and median ( range 133 . 39 ) , respectively . Similarly , in 560 datasets of size we estimated the following RSDs of TPR: 12 . 43 ( range 113 . 38 ) for C3D , 57 . 52 ( range 161 . 89 ) for WGCNA and 87 . 96 ( range 120 . 59 ) for DiffCoEx . The large RSDs of TPR calculated from the WGCNA and DiffCoEx analyzes originated because these methods often detected the simulated cluster ( s ) only in small number of replicates ( e . g . , 2 out of 20 ) . Besides , in a few cases the TP/FP rates of WGCNA and DiffCoEx were influenced by the adopted parameterization ( for instance , FPR in the WGCNA analysis of Cluster pattern 1 , Figure 3 ) , suggesting that different choices of the input parameters can affect the detection of clusters ( see Text S1 for additional details ) . The C3D algorithm is built on the HO-GSVD framework and as such does not require the user to specify ad-hoc parameters to detect common or differential clusters . In our implementation of the C3D algorithm the user can control the MER at a specified level before the cluster genes are empirically validated using a permutation-based procedure ( see Methods section ) . In these simulation studies , we have used two different MERs ( 5% and 20% ) to inform a suitable choice of MER that maximizes true positive without inflating false positive rates . On average , we observed a increase in the TPR when was adopted as compared with . However , we found no significantly higher FPR , which were always across all simulated datasets , this suggesting that using the less stringent in real data analyzes is likely to increase the detection of true gene clusters , without increasing significantly false positives . Finally , we used a standard desktop computer ( Mac Pro , GHz Quad-core Intel Xeon with 20 Gb RAM ) to evaluate the computational time required by C3D and compare it with WGCNA and DiffCoEx to analyze the simulated datasets . While the run time of C3D scales exponentially with the number of genes in the input matrices or the number of conditions , our Matlab implementation of C3D is relatively fast and requires only 1 , 200s to analyze a gene co-expression matrix in conditions and 10s to analyze a gene co-expression matrix in conditions ( Figure S2 ) . When compared with competing approaches , we assessed that to process simulated datasets of 1 , 000 and 10 , 000 genes ( with observations and conditions ) C3D requires significantly smaller CPU time than DiffCoEx ( up to 2 . 3 fold more CPU time ) and WGCNA ( up to 8 . 2 fold more CPU time ) , respectively ( Figure S2 ) . To show how C3D provides a powerful , practical framework for real genome-scale analyzes and yields new biological insights into pathways and molecular networks , we report an application to two large multi-tissue gene expression datasets in rats and humans . Transcriptional profiling was carried out by Affymetrix microarray in the rat and mRNA sequencing ( RNA-seq ) in humans , respectively . The microarray dataset consisted of genome-wide expression profiles ( probe sets ) that were measured in seven tissues ( adrenal , aorta , fat , kidney , left ventricle , liver and skeletal muscle ) in a panel of recombinant inbred rat strains [29] , which is a well characterized model of hypertension , metabolic syndrome and cardiovascular disease [27] , [30] , [31] . The RNA-seq datasets consisted of genome-wide transcriptomic data of human fetal neocortex , which have been generated to investigate the molecular mechanisms underlying differences in germinal zones of the developing human brain . The human dataset consisted of expressed genes which were analyzed in four regions of the fetal neocortex ( ventricular zone ( VZ ) , inner subventricular zone ( ISVZ ) , outer subventricular zone ( OSVZ ) and cortical plate ( CP ) ) from six 13–16 weeks postconception human fetuses [32] . In both rat and human analyzes , to identify common and differential clusters we extracted the top ten eigenvectors ( based on the modulus of the eigenvalues of the decomposition of ) as candidates which are then used as input for the cluster nodes selection and validation step of the C3D algorithm ( see Methods ) .
Building on the HO GSVD framework , we have developed a new algorithm ( C3D ) for efficient , parameter-free and automatic detection of co-expression clusters and networks in multiple conditions . Our method is designed for analysis of weighted ( and unweighted ) networks ( input matrices ) across conditions , enabling applications to diverse data types and structures . Although the original HO GSVD algorithm assumes the non-singularity of the co-expression matrix , by using the Moore-Penrose pseudo-inverse , our C3D algorithm can be applied to the non-invertible case . We show that when an exact HO-GSVD of the input matrices exists ( as defined in ( 4 ) , see Methods ) , our HO GSVD is able to extract the right decomposition basis through the eigen-decomposition of , whereas it finds an approximate decomposition of the data in the absence of an exact solution ( Figure S4 ) . In particular , our empirical simulations and real-case applications reveal that our approximate decomposition is able to capture both common and differential co-expression structures for a wide range of noise levels , suggesting that our algorithm can be useful for practical applications to genomic data . Here , through the HO GSVD of large-scale genomic datasets we aimed to uncover the complex interactions between genes ( networks ) that can occur within or across multiple conditions . One distinctive feature of our computational method is in the flexible and simultaneous identification of both “common” and “differential” sub-network structures across several conditions . Selecting informative vectors of , we provide different orderings of to reveal candidate clusters that are important to all conditions or specific to a sub-set of conditions; then , we can distinguish the specific conditions where the clusters are present using a permutation-based approach . This procedure allows to pinpoint automatically the specific conditions where the sub-network structures are present and , at the same time , to provide an empirical estimate of the statistical significance ( empirical P-value ) for each cluster identified . In simulation studies , we demonstrated how C3D outperforms competing approaches in accuracy and reliability while being computationally less demanding . We highlight how our method allowed accurate detection of clusters within complex structures ( i . e . , “common” , “nested” and “overlapping” networks ) by specifying only the desired level of statistical significance: misclassification error rate to assign genes to clusters and empirical P-value for cluster detection . In contrast with other approaches , C3D does not need the user to specify ad-hoc parameters related to the expected number of clusters or cluster density [15] or necessary to determine the optimal height cut-off in the gene clustering tree [13] , [16] , [17] . Typically , these unknown parameters need to be “finely tuned” on each dataset in order to obtain the best compromise between TP and FP for each cluster ( see Text S1 for additional details ) . We also showed that the results obtained by two competing and widely-used methods ( WGCNA and DiffCoEx ) were less stable than those provided by C3D . This was apparent in the significantly smaller relative standard deviations in TPR calculated across simulated datasets in the C3D analyzes as compared with WGCNA and DiffCoEx . Since C3D utilised raw gene expression data matrices as input , the higher stability of C3D might be due to the reduced influence of the small number of observations on the stability of co-expression estimates , which can result in extreme patterns of correlation changes , corresponding to stable and fragile co-expression , as previously shown [62] . The high stability in the results and the parameter-free “nature” of the HO GSVD approach make the C3D algorithm a powerful computational tool for real genomic data exploration and analysis . To demonstrate this point , we reported an application of C3D to two large transcriptional datasets: ( i ) microarray-based gene expression profiles in seven rat tissues and ( ii ) RNA-seq-based gene expression analysis of germinal zones from human fetal neocortex . In the rat analysis , we reported several functionally enriched co-expression clusters , including a previously identified inflammatory gene network driven by the IRF7 transcription factor that represents a gene expression signature of macrophages within complex tissues . While this co-expression network was experimentally validated [27] it was not recovered by WGCNA , that surprisingly placed the IRF7 transcription factor and many regulated target genes in the group of “non-clustered” genes . In addition , our C3D analyzes revealed novel gene co-expression networks in sub-sets of tissues . For instance , we identified a network comprising Hsp and known cardiomyopathy genes , which suggested coordinated regulation of heat shock proteins genes in multiple tissues , and their potential functional role in cardiovascular disease [50] . While this network was not recovered by either WGCNA or DiffCoEx analyzes , we were able to replicate this new finding using separate cardiac and liver gene expression datasets in humans ( Figure 4 ) . In the study of human fetal neocortex we demonstrated previously undescribed co-expression between cell cycle and ECM-receptor interaction pathways and support their role in the proliferation and self-renewal of neural progenitors . In addition , our analyzes highlighted that pathways central to later cognitive functions ( e . g . , calcium signaling , long-term potentiation , axon guidance ) are present at an early stage in the developing human brain [61] , which was not previously appreciated . These studies illustrated how our method can be effectively applied to leverage the vast stream of genome-scale transcriptional data that has risen exponentially over the last years , promising to aid the fine-scale characterization of both context-specific and systems-level networks and pathways .
In this step we assume the input data are non-square matrices , where the rows represent the observations and the columns indicate genes . The number of genes must be the same across datasets while the number of observations can differ . We first log transform the data and subtract for each gene its average gene expression to avoid capturing differences in average gene expression across conditions . We then calculate the co-expression matrices corresponding to each condition . Each represents the covariance matrix of the data in condition . As in classic principal component analysis , the columns of can be scaled to unit variance to work on the correlation matrices rather than the covariance . Alternatively , our algorithm can directly take any co-expression matrix as input . This feature of our algorithm allows to extract common and differential clusters from matrices based on different co-expression measures , including robust correlation ( e . g . Spearman , Kendall ) and non linear metrics such as mutual information [63] . Similarly to classic SVD , each observation from the input data can be characterized by its expression profile and represented by a data point in a dimensional space . The observations from all datasets are contained in a subspace of dimension , which thereafter is referred to as the HO GSVD subspace . Here , we aim at finding directions in the HO GSVD subspace that either capture the variability in gene expression that is common to all conditions ( common factors ) or that is specific to a subset of conditions ( differential factors ) . Inspired by [26] we developed a general algorithm that allows computation of an approximate solution to the HO GSVD problem in the non full column rank case . In the HO GSVD , are decomposed into where , is a diagonal matrix with elements for and contain the right basis vectors of the HO GSVD subspace where . The right basis vectors allow to identify set of genes ( clusters ) with similar co-expression patterns , that are either specific to a subset of conditions or common to all conditions . Here we explain the derivation of our HO GSVD-based algorithm in the general case of non-square matrices . The derivation and discussion of the special cases ( square , symmetric matrices with full rank and square , symmetric matrices with full rank ) is reported in Text S1 . In the most general case , we define the right basis vectors as the solution of the eigen-decomposition problem of the matrix ( 3 ) where is the arithmetic mean of all the pairwise quotients and denotes the Moore-Penrose inverse of the co-expression matrix [24] . Here the Moore-Penrose inverse is used as a substitute of since the invertibility of is not guaranteed when , which is the typical scenario in genomics . We now assume there is an approximate HO GSVD where is composed of orthonormal left basis vectors and . In this case , for all we have ( 4 ) and its Moore-Penrose inverse is given by ( 5 ) Therefore we have ( 6 ) since is full row rank . Hence we can rewrite as follows ( 7 ) When there exists a common subspace of dimension , with basis vectors , for which the decomposition of the co-expression matrices ( 4 ) is exact , equation ( 7 ) becomes an equality and the eigenvectors of will lead to the exact basis of the common subspace . In HO GSVD applications to genomics data , can be as large as the total number of observations ( i . e . , ) , and an exact common decomposition of the co-expression matrices might not be possible . In this case the eigenvectors of do not provide an exact decomposition of the subspace . Moreover , is not guaranteed to be non-defective and have a full set of real eigenvalues and eigenvectors . However , even in the absence of an exact common decomposition , the real part of the complex eigenvectors can be used to derive a low rank approximation of the common subspace and extract common and differential covariance structures from the data . To test the ability of our HO GSVD based algorithm to capture these covariance structures in the data in the presence of a “noisy” HO GSVD decomposition we performed an empirical simulation study ( see Text S1 for details ) . Our simulations suggest that if a common subspace of dimension with basis vectors explains a significant fraction of the variance in the original datasets , the approximation ( 4 ) holds and the first eigenvectors of the matrix ( corresponding to the largest eigenvalues of ) will provide a good approximation of the basis vectors of the HO GSVD subspace ( Figure S4 ) . We selected two large gene expression datasets from rats and humans , where genome-wide expression profiles were assessed in the same subject/animal across multiple tissues . The rat datasets consisted of microarray-based expression profiles for probe sets that were measured in adrenal , aorta , fat , kidney , left ventricle , liver and skeletal muscle tissues in a panel of recombinant inbred rat strains [29] . Microarray expression data were retrieved from ArrayExpress , http://www . ebi . ac . uk/arrayexpress/ , ( skeletal muscle , E-TABM-458; aorta , E-MTAB-322; liver , E-MTAB-323 , fat and kidney , E-AFMX-7; heart , MIMR-222; adrenal , E-TABM-457 ) ; gene expression summaries were derived using robust multichip average ( RMA ) algorithm [66] and normalized using Z-score transformation before analysis with C3D . The human data were retrieved from the Gene Expression Omnibus ( GEO ) database ( www . ncbi . nlm . nih . gov/geo ) under accession number GSE38805 . Briefly , total RNA from the VZ , ISVZ , OSVZ , and CP of six 13–16 wk postconception human fetuses was isolated from laser-capture microdissected Nissl-stained cryosections of dorsolateral telencephalon ( see [32] for additional details on experimental procedures ) . RNA-seq data were expressed as fragments per kilobase of exon per million fragments mapped ( FPKM ) values and normalized on log2 scale , yielding an expression matrix of in neocortex regions , which were analyzed by C3D . The Matlab implementation of the C3D algorithm , detailed instructions to run the code and an example of the simulated datasets used in these studies can be downloaded from http://www . csc . mrc . ac . uk/Research/Groups/IB/IntegrativeGenomicsMedicine/ contact information: enrico . petrettocsc . mrc . ac . uk or xiaolin . xiaocsc . mrc . ac . uk | Complex biological interactions and processes can be modelled as networks , for instance metabolic pathways or protein-protein interactions . The growing availability of large high-throughput data in several experimental conditions now permits the full-scale analysis of biological interactions and processes . However , no reliable and computationally efficient methods for simultaneous analysis of multiple large-scale interaction datasets ( networks ) have been developed to date . To overcome this shortcoming , we have developed a new computational framework that is parameter-free , computationally efficient and highly reliable . We showed how these distinctive properties make it a useful tool for real genomic data exploration and analyses . Indeed , in extensive simulation studies and real-data analyses we have demonstrated that our method outperformed existing approaches in terms of efficiency and , most importantly , reproducibility of the results . Beyond the computational advantages , we illustrated how our method can be effectively applied to leverage the vast stream of genome-scale transcriptional data that has risen exponentially over the last years . In contrast with existing approaches , using our method we were able to identify and replicate multi-tissue gene co-expression networks that were associated with specific functional processes relevant to phenotypic variation and disease in rats and humans . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] | [
"biotechnology",
"algorithms",
"systems",
"biology",
"computer",
"applications",
"computer",
"science",
"mathematics",
"algebra",
"genetics",
"applied",
"mathematics",
"computing",
"methods",
"biology",
"genomics",
"computational",
"biology",
"numerical",
"analysis"
] | 2014 | Multi-tissue Analysis of Co-expression Networks by Higher-Order Generalized Singular Value Decomposition Identifies Functionally Coherent Transcriptional Modules |
Human respiratory syncytial virus ( HRSV ) is the major cause of lower respiratory tract infections in children under 5 years of age and the elderly , causing annual disease outbreaks during the fall and winter . Multiple lineages of the HRSVA and HRSVB serotypes co-circulate within a single outbreak and display a strongly temporal pattern of genetic variation , with a replacement of dominant genotypes occurring during consecutive years . In the present study we utilized phylogenetic methods to detect and map sites subject to adaptive evolution in the G protein of HRSVA and HRSVB . A total of 29 and 23 amino acid sites were found to be putatively positively selected in HRSVA and HRSVB , respectively . Several of these sites defined genotypes and lineages within genotypes in both groups , and correlated well with epitopes previously described in group A . Remarkably , 18 of these positively selected tended to revert in time to a previous codon state , producing a “flip-flop” phylogenetic pattern . Such frequent evolutionary reversals in HRSV are indicative of a combination of frequent positive selection , reflecting the changing immune status of the human population , and a limited repertoire of functionally viable amino acids at specific amino acid sites .
Human respiratory syncytial virus ( HRSV ) is a leading cause of severe acute respiratory infection in childhood worldwide [1] and an important agent of acute respiratory infection in the elderly and immunocompromised [2] , [3] . Initial studies with monoclonal antibodies to the HRSV F and G proteins divided the virus into two major groups ( A and B ) [4] , [5] . Sequencing studies based on several HRSV genes have supported this major subdivision and lead to an additional genotypic classification , mainly based on the G protein gene , for epidemiological studies of HRSV . The genotypes of HRSVA and B show complex fluctuating dynamics , since they may co-circulate during a given season , with one or two dominant genotypes that are then replaced in consecutive years [6] , [7] , [8] , [9] , [10] , [11] . The G protein is a target for neutralizing antibodies , interacts with host cell receptors and is highly variable [12] , [13] , [14] , [15] . Most changes in the G protein are localized at an ectodomain containing two hyper-variable segments , separated by a highly conserved region between amino acids 164 and 176 , assumed to represent a receptor-binding site [15] . Experimental data show that the G protein is not required for virus infection in vitro under appropriate conditions , but is necessary for efficient infection in mice and humans [16] . It has been argued that the antigenic variability of HRSV strains is one of the key features contributing to the ability of the virus to re-infect people and cause large-scale yearly outbreaks [17] . Moreover , several studies have shown that the C-terminal hyper-variable region of the surface G glycoprotein is immunogenic and contains multiple epitopes that are recognized by both murine monoclonal antibodies and human convalescent sera [18] . In addition , the deduced amino acid sequences of the G protein are highly divergent , with a sequence identity of approximately 53% between HRSVA and B , and 20% divergence within the same antigenic group [14] , [19] . Despite this diversity , the nature of the selection pressures acting on the G protein have not been explored in detail , and particularly using sequence data sets that are of sufficient size to reveal the intricate nature of adaptive evolution and with restricted spatial and temporal sampling . This information is of particular importance since the ectodomain of the G-protein is also a target site in vaccines that have so far met with little success . To addresses these key issues we undertook the largest analysis of HRSV sequences undertaken to date , comprising both HRSVA and HRSVB , and utilizing detailed temporal information .
A total of 3 , 496 respiratory samples were used in this study . Nasopharyngeal aspirates and nasal swabs from 2 , 256 infants and young children ( 1 week to 5 years of age ) hospitalized with acute respiratory lower infection ( ARI ) at University of São Paulo Hospital , São Paulo , Brazil were used . Samples were collected over 11 consecutive HRSV seasons ( 1995–2005 ) . In addition , 1 , 240 respiratory samples from children with ARI were collected in 2004 and 2005 in different cities in São Paulo State and enrolled in the present study as part of the Viral Genetic Diversity Network ( VGDN ) ( http://www . lemb . icb . usp . br/LEMB/index . php ? p=11 ) . Informed consent was obtained from parents or guardians of children enrolled in the study in the different cities according to a protocol approved by their respective Institutional Review Boards . Specimens were collected in buffered saline and transported on ice to the laboratory for processing within 4 hours . A commercial immunofluorescence assay was used per manufacturer's instructions ( Chemicon Light Diagnostics , Millipore Corp , Inc . , Temecula , CA . ) , as previously described [20] . Clinical samples were amplified by RT-PCR as described bellow . Total RNA was extracted using guanidinium isothiocyanate phenol ( Trizol LS , Invitrogen® , Carlsbad , CA ) according to the manufacturer's instructions . Extracted RNA was annealed with 50 pmol random hexanucleotide primer ( Invitrogen® ) at 25°C for 25 minutes , followed by reverse transcription with 200 U SuperScript ™ ( Invitrogen® ) at 42°C for 1 hour . Partial HRSV G gene amplification was performed by a semi-nested PCR procedure . cDNA was amplified with reverse primer FV -5′GTTATGACACTGGTATACCAACC 3′ ( based on sequences complementary to nucleotides 186 to 163 of the F protein gene messenger RNA strain CH18537 [21] – and the forward primer GAB - 5′YCAYTTTGAAGTGTTCAACTT 3′ ( G gene , 504–524 nt ) . A semi-nested PCR was then performed with primers F1AB -5′CAACTCCATTGTTATTTGCC3′ ( F gene , 3–22 nt ) and GAB [8] , [9] . PCR assay was carried out in a reaction mixture containing 2 , 5 µL of cDNA , 20 mM Tris-HCl , 50 mM KCl , 1 , 5 mM MgCl2 , 0 , 2 mM dNTPs , 10 pmol of each primer , 1 , 25 U of Taq DNA Polimerase ( Taq-Gold , Applied Biosystems Inc ) in a final volume of 25 µL . Amplification was performed in a GeneAmp PCR System 9700 thermocycler ( Applied Biosystems Inc . ) with the following parameters: 94°C for 5 minutes , followed by 35 cycles of 1 min at 94°C , 1 min at 55°C and 1 min at 72°C , and finally 7 min of extension at 72°C . The semi nested PCR was carried under the same conditions , with 10 pmol of each primer on a final volume of 25 µL . Both cDNA synthesis and PCR followed strict procedures to prevent contamination , including redundant negative controls and segregated environments for pre- and post-amplification procedures . Amplified products of the G gene showing the expected size by gel electrophoresis , were purified with a commercial kit ( Concert Gel Extraction Systems , Invitrogen® ) , according to the manufacturer's instructions , followed by cycle-sequencing on a GeneAmp PCR System 9700 thermocycler ( Applied Biosystems Inc ) . Sequence reactions were subjected to electrophoretic separation for primary data collection in ABI PRISM 3100 and 377 DNA sequencers ( Applied Biosystems Inc . ) , using a fluorescent dye terminator kit ( Applied Biosystems Inc . ) . Both strands were sequenced at least twice . Sequences were assembled with the Sequence Navigator program version 1 . 0 ( Applied Biosystems Inc . , EUA ) resulting in contigs of 270 nucleotides on average , corresponding to HRSV G gene nucleotide 649–918 ( group A , prototype strain A2 ) and 652–921 ( group B , prototype strain CH18537 ) . Individual sequences were aligned to HRSV G references sampled globally [8] , [9] , [10] , [11] , [19] , [22] , [23] , [24] , [25] , [26] , [27] , [28] , [29] , [30] , [31] , [32] , [33] , [34] , [35] , [36] , [37] with the Se-Al - Sequence Alignment Editor [38] , resulting in data sets with an average length of 270 nucleotides . Because the strains A2 and Long ( A group ) and CH18537 and Sw8/60 ( B group ) were both the most divergent and considered prototype strains , they were included in the analysis as outgroup sequences for the phylogenetic analysis . The best–fit model of nucleotide substitution ( GTR+Γ+I ) , and values for the shape parameter ( α ) for the distribution of among-site rate-heterogeneity distribution ( Γ ) were selected by hierarchical likelihood ratio testing using Modeltest Version 3 . 06 [39] . Using these models , maximum likelihood ( ML ) phylogenetic trees were inferred by heuristic searches using PAUP under sequentially the TBR ( Tree Bisection-Reconnection ) , SPR ( Subtree Pruning Regrafting ) and NNI ( Nearest Neighbor Interchange ) perturbation procedures [40] , using as BioNJ tree as a starting phylogeny . Levels of phylogenetic support for individual nodes were obtained by obtaining the majority rule consensus of the 100 best trees collected near the likelihood maxima during both the SPR and NNI branch-swapping procedures . Consensus values above 99% were used to check for genotype monophyly . Moreover , since samples had dates of sampling ranging a 30 year period , we also generated maximum clade credibility ( MCC ) trees for HRSVA and HRVB data using the Bayesian inference ( BI ) method in BEAST v . 1 . 4 . 7 [41] . We used the best fit model ( GTR+Γ4+I ) assuming an uncorrelated lognormal-distributed relaxed clock with rates of change estimated from the data and using a Bayesian skyline demographic model as a coalescent prior . To obtain effective sampling sizes ( ESS ) above 100 , MCC trees for HRSVA and HRSVB were obtained by pooling five independent Markov-chain Monte Carlo runs , each of which sampled from 10 million chains after a pre-burning period of 30 million chains . To detect sites in the G protein that might be subject to positive selection we used the Bayesian methods implemented in the HyPhy program [42] . We employed the default ‘MG94xHKY85x3_4x2Rates with Rate heterogeneity , with 4 rate categories per parameter’ model . This estimates multiple parameters that are free to vary over sites both dN and dS to have distinct rates at a given site and to be sampled independently from two separate distributions . For the MG94xHKY85x3_4x2Rates model we used a Bayes factors >20 means that positive selection explains the data approximately 20 times better than the alternative model [42] . For comparison , we also used the less computationally intensive Single Likelihood Ancestral Counting ( SLAC ) and Fixed-Effects Likelihood ( FEL ) methods using the best fit nucleotide model estimated with HyPhy for each data set . With SLAC and FEL , all positively selected sites were estimated at the 95% confidence interval . We did not use the random effects likelihood method ( REL ) because of major computational constraints [42] , [43] , [44] . In addition , we obtained the most parsimonious reconstructions ( MPR ) of the positively selected sites along both HRSVA and HRSVB G protein trees using both the ‘accelerated transformation’ ( ACCTRAN ) method that maps character changes near the root of the tree , and the ‘delayed transformation’ ( DELTRAN ) method that maps character changes near the tips of the tree implemented in MacClade v . 4 . 07 [45] . Because all sequences had dates of sampling , allowing us to recover temporal patterns of amino acid replacement , we adjusted the tips of the phylogenies in time ( i . e . , tip-dated trees ) with BEAST v . 1 . 4 . 7 ( http://beast . bio . ed . ac . uk/ ) . The nucleotide sequences from the Brazilian isolates were deposited in the GenBank database under accession numbers EU582054 to EU582483 , EU635778 to EU635865 , EU259652 to EU259673 , EU259675 , EU259676 , EU259678 to EU259690 , EU259693 to EU259696 to EU259704 , EU 625735 , EU241632 to EU241634 and AY654589 .
We obtained nucleotide sequences of the second region ( G2 ) of the HRSV G protein gene from ( i ) 432 random samples collected over 11 seasons ( i . e . , from 1995 to 2005 ) from the city of São Paulo , Brazil and ( ii ) 136 sequences from samples collected by VGDN program from 2004 to 2005 from metropolitan area of the city of São Paulo and from the city of Ribeirão Preto , also in São Paulo state . Of a total of 568 sequences , 359 ( 63 . 2% ) represented group A and 209 ( 36 . 8% ) group B . The Brazilian isolates of HSRVA had a deduced G protein of 298 or 297 amino acids , which was confirmed by complete G protein sequences obtained from several representative samples used in this study ( data not shown ) . Interestingly , three isolates ( Br89_2000 , Br86_2000 and Br 206_2004 ) had a premature stop codon at amino acid position 288 , which has been observed previously [46] , [47] and four Brazilian GA5 strains from 2000 had a deletion of three bases , causing the loss of a serine residue at position 270 . All the Brazilian isolates of HSRVB had an inferred G protein of 295 amino acids , except three isolates from 2000 season that had 299 amino acids due to a mutation in the first nucleotide of the stop codon of G gene , and two strains from 1999 season that had a Threonine codon insertion at position 233 , leading to a G protein with 296 amino acids . Some strains isolated during 2001 , 2003 and 2004 have an exact duplication of 60 nucleotides starting after residue 791 ( accounting for a 20 aa duplication by insertion and resulting in a predicted protein of 312 to 319 aa ) . In 2005 this new genotype – denoted ‘GB3 with insertion’ ( see below ) – became the predominant . The alignment of partial amino acid sequences including the duplicated G segment showed some amino acid substitutions in the duplicate segment and in the 60 nucleotides immediately upstream . A total of 933 HRSVA sequences and 673 HRSVB sequences , including original and data compiled from GenBank , were used for further evolutionary analysis ( see Table S1 in supplementary material ) . Both the ML and MCC trees divided HRSVA into seven monophyletic clusters with bootstrap or posterior probability support above 99%; these were previously described as genotypes GA1 , GA2 , GA3 , GA4 , GA6 , GA5 , GA7 and SAA1 [8] , [9] , [34] . The MCC tree ( Fig . 1 ) showed that genotypes GA2 , GA3 , GA4 , GA6 and GA7 had a common ancestor not shared by GA5 and GA1 . GA2 strains fell into two distinct branches . One included the oldest strains isolated from 1995 to 2000 in several regions globally ( i . e . , Brazil , South America , Belgian , United States and Africa ) . The other branch grouped the most recently isolated strains ( 2000 to 2004 ) that also exhibited a very widespread distribution ( i . e . , Belgian , Brazil , China and Africa ) . Brazilian strains in this second branch were characterized by five amino acid substitutions: Leu215Pro , Arg244Lis , His266Tyr , Asp297Lys and the stop codon at 298 reverting to Trp ( Stop298Trp ) . These changes were fixed in almost all 2003 to 2005 GA2 strains . The MCC tree for HRSVB ( Fig . 2 ) contained 8 clusters that were previously described as genotypes JA1 , GB1 , GB2 , GB3 , GB4 , SAB1 , SAB2 , SAB3 and GB3 with insertion [8] , [9] , [33] , [34] . These same groupings were observed in the ML tree . GB3 was a paraphyletic genotype and included both SAB3 and the GB3 with the 60 nucleotide insertion . In total , we found 29 sites to be subject to elevated rates of non-synonymous substitution ( dN ) in HRSVA , nine of which were detected to be under positive selection by the three methods we used and strongly suggesting that they are not false positives ( Table 1 ) . By using the Long 1956 prototype strain as an outgroup sequence , the most parsimonious reconstruction of the positively selected amino acid changes along the HRSVA tip-dated tree revealed that 22 mapped to the basal node ( Fig . 1 ) . That these 22 putatively positively selected sites included the replacement substitutions that defined the different genotypes and lineages within genotypes confirmed that they have reached a high frequency in the population again as expected of bona fide positively selected sites rather than false-positives . Site 215 had a Leucine ( Leu ) in all genotypes except in the non-circulating genotype GA1 , that had a Proline ( Pro ) , and the non-circulating Long 1956 prototype strain , that possessed a Histidine ( His ) . Interestingly , most GA2 strains isolated after 2000 reverted to Pro at this site , and most GA5 strains isolate since 2001 changed to Isoleucine ( Ile ) . Moreover , GA1 was basal to all other HRSVA genotypes and since it had no reversals on positive selected sites it was excluded from Fig . 1 for the sake of clarity ( Fig . 1 ) . The changes Phe265Leu , Leu274Pro , Ser280Tyr , Pro286Leu , Ser290Pro and Pro293Ser mapped to the split of two distinct branches; ( i ) one containing the older samples including prototypes ( A2 and Long ) and non-circulating genotype GA1 and , ( ii ) another containing the remaining genotypes ( GA5 , GA2 , GA3 , GA4 , GA6 and GA7 ) ( Fig . 1 ) . Val225Ala , Pro256Leu , Thr238Leu and Leu274Thr changes defined the GA5 genotype ( Fig . 1 ) . Moreover , positively selected changes Pro289Ser , Pro226Leu , Ser269Thr and Pro290 Leu defined GA2 genotype , while Pro226Leu defined genotypes GA3 and GA7 ( Fig . 1 ) . A total of 23 sites had elevated non-synonymous rates ( dN ) in HRSVB , thirteen of which were detected to be under positive selection by the three methods we used ( Table 2 ) . As with HRSVA , these sites defined lineages within genotypes , again suggesting that they are not false-positive results . The substitutions Pro216Ser and Pro219Ser defined the GB4 genotype , while changes Leu237Pro and Pro219Leu defined genotype GB3 , and Thr255Ala defined SAB3 genotype . Moreover , site 277 changed from Ser to Phe in genotype JA1 ( isolated in Japan ) and in the SAB1 genotype . Finally , positively selected sites 242 , 247 , 255 , 257 and 258 were located immediately upstream of the 20 amino acid-long duplication region while sites 267 , 269 and 270 were located at the duplication region of the gene . The MPR of positively selected amino acid changes along the HRSVB tip-dated tree revealed that sites 222 , 227 , 255 , 257 , 276 , 291 and 293 were likewise associated with the split of two distinct branches; ( i ) one containing the older samples including prototypes ( CH18537 and Sw860 ) and non-circulating genotype JA1 and , ( ii ) other containing the remaining genotypes ( SAB1 , SAB2 , SAB3 and GB3 ) ( Fig . 2 ) . However , perhaps the most notable observation of this analysis was that 18 of the total of the 55 putatively positively selected sites in HSRVA and B tended to revert , in time , to a previous codon state , indicative of a reversible ( i . e . , “flip-flop” ) pattern of amino acid replacement ( shown in bold in Table 1 and 2 , Fig . 1 and 2 ) . Eleven of these 18 reversible sites in RSVA and RSVB were found to be positive selected under the most sensitive model ( MG94xHKY85x3_4x2Rates model ) but most by more than one models and by at least one model . Strikingly , such reversible evolution occurred at nine sites independently in both HRSVA and in HRSVB , although two sites in each virus group experienced reversal without detectable positive selection . For example , site 290 in GA2 genotype reverted from Leu to Pro five times along the tree ( Table 1 and Fig . 1 and 3 ) . Similarly , in some HRSV B genotypes ( GB3 with insertion , GB3 and SAB3 ) site 219 reverted from Leu to Pro seven times along the tree ( Table 2 , Fig . 2 and 4 ) .
The Brazilian isolates of G protein sequences of HRSV A and B demonstrated remarkable genetic flexibility , as noted previously at the global scale [33] , [48] , [49] , [50] , [51] , [52] . Such a high level of genetic variation may be associated with the fact the G protein plays a key role in facilitating reinfections in HRSV – allowing evasion from cross-protective immune responses – and hence in the fluctuating patterns of viral circulation . As a consequence , describing the complex patterns of amino acid change in both HRSVA and HRSVB over time may help understand the evolution and epidemiology of this important virus . Our analysis revealed that the ectodomain of the G protein was subject to strong positive selection , with 29 positively selected amino acid sites in HRSVA and 23 amino acid sites in HRSVB . The action of positive selection at these sites was also strongly supported since 18 of the 52 putatively positive selected sites were detected using all three forms of dN/dS analysis . Only 5 of the 29 positively-selected sites in HRSVA have described previously ( 215 , 225 , 226 , 256 , 274 and 290 ) [10] , [53] , [54] . Possibly , this difference is due to the far larger data set available here and/or use of different analytical methods . Further , many of the positively selected sites in group A defined genotypes and lineages within genotypes , and correlated well with known epitopes described in escape-mutants selected with specific Mabs ( sites 226 , 237 , 265 , 274 , 275 , 284 , 286 and 290 ) [25] , [49] , [51] or in natural isolates ( sites 215 , 225 , 226 , 265 , 280 , and 293 ) [13] , [25] , [46] . It is interesting to note that three of these sites ( 226 , 265 , 290 ) defined genotypes and underwent frequent reversals ( Fig . 1 ) . Site 237 was unique among positively selected sites in group A in that it had a residue – Asn – with the potential for N-glycosylation [14] . Moreover , six positively selected sites ( 225 , 227 , 253 , 269 , 275 , 287 ) were previously described to have O-linked side chains [1] . The frequency and pattern of glycosylation were important in defining the antigenicity of the G protein , either by masking antigenic sites or by recognition of specific antibodies [55] , [56] . Less is known about the effects of amino acid replacements at other sites ( 222 , 227 , 230 , 243 , 246 , 248 , 249 , 272 , 279 , 285 and 292 ) , although they were located close together to some of the epitopes involving in neutralizing the virus ( Fig . 3 ) . Moreover , we observed differences in the length of the G protein due to a stop codon mutation at site 298 , and which was associated with the split of the tree in different branches . In GA5 , Gln298 was maintained but changed to a stop codon ( Gln298Stop ) in both GA1 and the lineages leading to all other genotypes GA2 , GA3 , GA4 , GA6 and GA7 . Interestingly , the stop codon at 298 reverted to Trp in the GA2 branch that contains the most recent isolates . This reflects amino aid replacements involved in the presentation or elimination of multiple epitopes containing the three last residues of the G protein ( i . e . , 296 to 298 C-terminal ) [57] . Although epitopes in HRSVB are not well characterized , important differences in protein length between Brazilian strains were observed ( 295 or 299 amino acids ) , due to differences in the occurrence of the final stop codon ( site 293 ) . It was suggested that this region presents an epitope , substitutions in which would abolish the recognition of the G protein by strain specific antibodies [50] , [54] . Moreover , the change at site 293 ( stop codon:Gln ) divided the tree in two distinct branches , one that included the ancient non-circulating strains and the other that included the recent strains . In almost all the Brazilian HRSV GB3 with insertion strains isolated in 2005 we observed an evolutionary “flip-flop” between a glutamine at site 293 and a stop codon , leading to a predicted G protein of 312 amino acid in length . Remarkably , other sites experienced similar reversals , such as amino acids 219 , 227 , 237 and 257 , which defined new genotypes , suggesting that there are a limited number of amino acid residues at this site that allow successful virus attachment glycoprotein . Indeed , HRSV escape mutants that differ in their last 81 residues from the canonical Long prototype protein sequence , retain their compositions and hydropathy profiles [25] , strongly suggesting that there may be indeed structural restrictions to changes in the G protein , although this will need to be investigated further . Finally , positively selected sites located in the 20 amino acid duplicated region of the gene , and immediately upstream of it may influence the expression of some important epitopes . For example , the additional O-linked glycosylation residues in both the insertion and duplication regions probably confers advantage of this new variant over the other HRSVB genotypes . Of the 23 positively selected sites in HRSVB , only five were described previously by Zlateva et al . 2005 ( sites 219 , 237 , 247 , 257 and 258 ) and , two by Woelk and Holmes , 2001 ( sites 227 and 257 ) . Consequently , HRSV appears to be subject to far greater positive selection pressure than previously realized . Our data also identified amino acid sites under positive selection sharing positional homology in the two groups . For example , 11 positively selected sites in HRSVA ( 215 , 226 , 246 , 256 , 265 , 274 , 275 , 284 , 285 , 290 and 292 ) had positional homologues in HRSVB ( 216 , 227 , 247 , 257 , 266 , 275 , 276 , 285 , 286 , 291 and 293 ) . Some of these sites are known to harbor epitopes in HRSVA ( 215 , 226 , 265 , 275 , 284 and 290 ) . Moreover , some sites were important in defining lineages in the phylogenetic tree , such as sites 215 , 265 , 274 , 286 and 290 , specific to prototypes and non-circulating GA1 genotypes and site 226 defining the GA2 genotype . Less is known about epitopes in HRSVB , but sites 227 , 257 , 276 , 291 and 293 , under positive selection , were associated with the major division of the HRSVB phylogenetic tree into two branches . The most interesting observation from this analysis was that both HRSVA and HRSVB experienced frequent evolutionary reversals of amino acids at positively-selected sites Tables 1 and 2 , Figs . 1 and 2 ) , which in turn mapped to known and possibly newly-described epitopes ( Figs . 3 and 4 ) . That most of the sites experiencing this “flip-flop” evolutionary pattern were also under positive selection strongly suggests that they reflect the fluctuating dynamics in the immune status of human populations , in which patterns of cross-protective immunity ebb and wane . To be more specific , the build-up of lineage-specific resistance in the host population would drive the process of positive selection in key immunological epitopes . Later , following the loss of herd immunity to the previous viral epitope , coupled with constraints which mean that only a limited number of amino acids are functionally viable , a reversion mutation would be fixed by positive selection in a newly susceptible human population . In sum , the frequent evolutionary reversals observed in the G protein of HRSV are a necessary consequence of a limited set of possible replacements at HRSV epitopes . Without such a constraint on the repertoire of functionally viable amino acids we would expect to see a gradual diversification at these sites rather than frequent reversals . This model agrees well with the spacing of temporal events observed in both viral phylogenies , supporting the notion that reversible evolution may contribute to the escape from the human population immune response , thereby facilitating viral transmission . A clearer understanding of the determinants of the evolutionary reversals within the G protein could ultimately lead to a better understanding of the viral immune-escape repertoire and assist in the control of HRSV . | As part of the Viral Genetic Diversity Network ( VGDN ) , we sequenced the second variable region ( G2 ) of the G protein of human respiratory syncytial virus ( HRSV ) A and B from 568 patients sampled during 11 consecutive HRSV seasons ( 1995–2005 ) in the state of São Paulo , Brazil . A total of 933 HRSVA and 673 HRSB time-stamped sequences , including those sampled here and globally , was used for phylogenetic inference and the analysis of selection pressures . We identified 18 positively selected sites in both HRSVA ( 9 sites ) and HRSVB ( 9 sites ) that tended to revert in time to their previous codon state ( i . e . exhibited a “flip-flop” pattern ) . We argue that these common evolutionary reversals are indicative of frequent positive selection , reflecting the changing immune status of the human population , coupled with a limited repertoire of functional viable amino acids at specific sites . This information is of particular importance since the ectodomain of the G protein is also a target site in vaccines that have so far proven unsuccessful and because it constitutes a significant step towards describing and understanding the immune-escape repertoire of this major human pathogen . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [
"virology/immune",
"evasion",
"evolutionary",
"biology/microbial",
"evolution",
"and",
"genomics",
"virology/virus",
"evolution",
"and",
"symbiosis",
"evolutionary",
"biology/evolutionary",
"and",
"comparative",
"genetics"
] | 2009 | Positive Selection Results in Frequent Reversible Amino Acid Replacements in the G Protein Gene of Human Respiratory Syncytial Virus |
Kaposi's sarcoma-associated herpesvirus ( KSHV ) is etiologically related to Kaposi's sarcoma ( KS ) , primary effusion lymphoma ( PEL ) and multicentric Castleman's disease ( MCD ) . It typically displays two different phases in its life cycle , the default latency and occasional lytic replication . The epigenetic modifications are thought to determine the fate of KSHV infection . Previous studies elegantly depicted epigenetic landscape of latent viral genome in in vitro cell culture systems . However , the physiologically relevant scenario in clinical KS tissue samples is unclear . In the present study , we established a protocol of ChIP-Seq for clinical KS tissue samples and mapped out the epigenetic landscape of KSHV genome in classic KS tissues . We examined AcH3 and H3K27me3 histone modifications on KSHV genome , as well as the genome-wide binding sites of latency associated nuclear antigen ( LANA ) . Our results demonstrated that the enriched AcH3 was mainly restricted at latent locus while H3K27me3 was widespread on KSHV genome in classic KS tissues . The epigenetic landscape at the region of vIRF3 gene confirmed its silenced state in KS tissues . Meanwhile , the abundant enrichment of LANA at the terminal repeat ( TR ) region was also validated in the classic KS tissues , however , different LANA binding sites were observed on the host genome . Furthermore , we verified the histone modifications by ChIP-qPCR and found the dominant repressive H3K27me3 at the promoter region of replication and transcription activator ( RTA ) in classic KS tissues . Intriguingly , we found that the TR region in classic KS tissues was lacking in AcH3 histone modifications . These data now established the epigenetic landscape of KSHV genome in classic KS tissues , which provides new insights for understanding KSHV epigenetics and pathogenesis .
Kaposi's sarcoma-associated herpesvirus ( KSHV ) was first identified in Kaposi's sarcoma ( KS ) biopsies by Chang and Moore in 1994 and has been proven to be the etiological agent of several human cancers including KS , primary effusion lymphoma ( PEL ) and multicentric Castleman's disease ( MCD ) [1–3] . KSHV is a double stranded DNA virus with a large genome about 170 Kb , belonging to the gamma herpesvirus subfamily [4 , 5] . It typically displays two different phases in its life cycle , the default latency and occasional lytic replication [5] . During latency , the viral genomes persist as episomes with limited latent gene expression in the nucleus of the infected cell and no virion is produced [6 , 7] . The latent genes are grouped at one locus in the genome , including ORF73 ( latency-associated nuclear antigen , LANA ) , ORF72 ( vCyclin ) , ORF71/K13 ( vFlip ) , K12 ( Kaposin ) and a cluster of miRNAs [8 , 9] . Under specific conditions such as hypoxia , cell stress and valproic acid or butyrate stimulation , the virus will reactivate from latency with an orchestrated expression of lytic genes , leading to the massive production of mature virions [10–12] . Replication and transcription activator ( RTA ) encoded by ORF50 is the key switch regulator that controls KSHV reactivation [13 , 14] . Adding inhibitors of DNA methyltransferases or histone deacetylases to KSHV infected cells can effectively induce the expression of RTA , which promotes the virus entering lytic replication from latency [11 , 15 , 16] . Since the epigenetic modifications are thought to determine the fate of KSHV infection , it is important to understand the epigenetic status of viral genome during latency and reactivation [17–20] . Previous studies have elegantly depicted the genome-wide histone modifications on KSHV genome in in vitro cell culture systems [21 , 22] . It has been demonstrated that activating histone modifications like acetylation of histones ( AcH3 ) are only enriched in several loci while repressive histone modifications like H3K27me3 are widespread across the viral genome , which well explained the expression pattern of viral genes during latency [21 , 22] . While most studies are established on the usage of in vitro cultured PEL and KSHV-infected endothelial cell lines , the physiologically relevant scenario in clinical KS samples is unclear . KS is a highly vascular sarcoma on the skin originated from endothelial cells , which is characterized by the infiltrated inflammatory cells and neo-angiogenesis [8 , 23] . According to the geographical distribution and clinical outcomes , KS can be classified into four subtypes , which are classic , endemic , iatrogenic and AIDS-related KS . All the subtypes of KS lesions share a common histological characteristic but are substantially different in disease progression [23–26] . The expression pattern of latent genes in KS is not exactly the same as the one in PEL [27 , 28] . A previous study has shown that the expression of vIRF3 gene is only detected in PEL samples , which suggests a distinct epigenetic status in KS [28] . Typically , cells derived from KS tissues will loss episomes very quickly during the in vitro culture , thus it is difficult to obtain cell lines reflecting the physiologically relevant scenario in KS tissues [29 , 30] . Therefore , it is important to directly determine the epigenetic landscape of KSHV genome in KS tissues . In the present study , we established a protocol of ChIP-Seq for clinical KS samples and directly mapped out the epigenetic landscape of KSHV genome in classic KS tissues which are only associated with KSHV infection and derived from Xinjiang area of China . Specifically , we examined AcH3 and H3K27me3 histone modifications on KSHV genome , as well as the genome-wide LANA binding sites . Our results demonstrated that the enriched AcH3 histone modifications were mainly restricted at latent locus while H3K27me3 histone modifications were widespread on KSHV genome in classic KS tissues . The epigenetic landscape at the region of vIRF3 gene confirmed its silenced gene expression in KS tissues . Meanwhile , the abundant enrichment of LANA at the terminal repeat ( TR ) region was also validated in the classic KS tissues , however , different LANA binding sites were observed on the host genome . Furthermore , the dominant repressive H3K27me3 histone modifications at RTA promoter region were verified by ChIP-qPCR . Intriguingly , we found that the TR region in classic KS tissues was lacking in AcH3 histone modifications with abundant LANA accumulation . Moreover , the established epigenetic landscape in KS tissues was further confirmed in new cases of classic and AIDS-related KS tissues . By analyzing histone modifications and LANA binding sites in classic KS tissues , our study provides new insights for the understanding of KSHV epigentics .
Previous studies on the epigenetic landscape of KSHV genome using in vitro cell culture systems have systemically determined genome-wide distributions of four well-known histone modifications by ChIP-on-ChIP , including AcH3 , H3K4me3 , H3K27me3 and H3K9me3 [21 , 22] . These results supported the predicted activating role of AcH3 and predominant repressive role of H3K27me3 on the KSHV genome . To obtain a physiologically relevant map of the epigenetic landscape of KSHV genome in classic KS tissues , we performed ChIP-Seq experiments on classic KS tissues originated from two different patients . The specific protocol of ChIP in clinical KS tissues was summarized in the Fig 1 and material and methods section . Each experiment was divided into five experimental groups which are input , IgG , LANA , AcH3 and H3K27me3 . The purified ChIP product from input , LANA , AcH3 and H3K27me3 groups were subjected to next generation-sequencing . Sequence reads for each sample were aligned to the KSHV genome ( HQ404500+35TR ) and human genome ( Hg19 ) using Bowtie2 [31] . The results of alignment was presented in Table 1 . The overall alignment rate to the KSHV genome was around 0 . 02% . The relatively high rate in the H3K27me3 group suggested a possible enrichment in KSHV genome . The aligned files were subjected to peak calling and generation of genome-wide maps by using Model-based Analysis of ChIP-Seq ( MACS ) and Hypergeometric Optimization of Motif EnRichment ( Homer ) software [32 , 33] . The general maps of AcH3 and H3K27 histone modifications on the KSHV genome are illustrated in Fig 2 . The enlarged maps are presented in Fig 3 ( AcH3 ) and Fig 4 ( H3K27me3 ) . The peak panels illustrated in Figs 2 , 3 and 4 demonstrates the most potentially and significantly enriched signals on the KSHV genome . As shown in Fig 2 , the enriched AcH3 histone modifications were mainly restricted to the latent locus while H3K27me3 histone modifications were widespread on the KSHV genome . The comparison between AcH3 and H3K27me3 panels showed mutually exclusive signals ( high AcH3 level with low H3K27me3 level ) on the latent locus and several loci , including the promoter region of vIRF3 gene , regions around ORF8 gene and K5 gene . The coding region of the vIRF3 gene was dominated by the repressive H3K27me3 histone modifications ( Figs 3 and 4 ) , which suggests that the vIRF3 gene is silenced but may be easily activated in classic KS tissues . Meanwhile , we also validated that the expression of vIRF3 was restricted at extremely low level in classic KS samples ( S1 Fig ) . This result was in line with previous findings that the vIRF3 expression is not detected in KS tissues by immunohistochemistry analysis [28] . The epigenetic landscape of the KSHV genome in these two cases of classic KS tissues were different from each other in several loci . The first case showed higher level of AcH3 histone modifications on the KSHV genome than the second case in general ( Figs 2 and 3 ) , although there was similar trend . To be noted , the promoter and coding regions of the K15 gene were enriched with AcH3 histone modifications only in the first case ( Fig 3 ) . In the meantime , we observed a unique peak of H3K27me3 at the promoter region of the LANA gene only in the first case ( Fig 4 ) . The higher level of AcH3 and the unique peak of H3K27me3 at the promoter region of the LANA gene in the first case provides supporting evidence for the repressive role of LANA on viral gene expression as previously reported [34–37] . The difference in the overall epigenetic landscape between these two cases indicates different states of KSHV infection in these two patients , which might have clinical relevance to KS progression ( S1 File ) . By comparing with previously published epigenetic maps of the KSHV genome [21 , 22] , we found that the enriched AcH3 signals at multiple loci ( e . g . 10 Kb; 20–30 Kb; 87 Kb ) in in vitro cell culture systems ( BCBL-1 and KSHV infected SLK ) were not observed in classic KS tissues while the landscape of H3K27me3 histone modifications in in vitro cell culture systems was much similar to the one in KS tissues . It has been reported that KSHV may have different latency programs in different tissues or cell lines and the expression pattern of viral genes could be affected by the cytokines present in the local cellular milieu [27 , 38] . The difference in AcH3 histone modification of classic KS tissues might indicate a relatively mild environment with less cytokines for classic KS tissues . Files for the generation of genome-wide maps were provided in the S1 Supporting Information section . LANA protein is critical for the maintenance of KSHV episome [39–41] . The genome-wide LANA binding sites in the in vitro cell culture systems were well described in several studies [42–47] . Yet its footprint is not known in KS tissues which also shows consistent LANA expression . Therefore , it is important to investigate the behavior of LANA in KS tissues . We designed the experimental group of LANA in the ChIP-Seq experiments . The genome-wide LANA binding sites on the KSHV genome in classic KS tissues are illustrated in Fig 5A . The abundant enrichment of LANA at the terminal repeat ( TR ) region was validated in both of the two cases of classic KS tissues whereas previously reported the enrichments at the latent locus of LANA was only found in the first case , which further confirmed different states of KSHV infection in these two patients . The panels presented several small peaks across the genome , but was not consistent in these two cases , and was difficult to distinguish from the background noise . Although several weak binding sites of LANA were found on KSHV genome in addition to the TR region and latent locus in the previous studies [42–47] , the results in classic KS tissues did not show these peaks . This difference may be a result of the reduced sensitivity of ChIP-Seq for the weaker protein binding sites and the smaller size of classic KS tissues . In the meantime , we also analyzed LANA binding sites on the host genome . The identified peaks of LANA binding by MACS were subjected to Peak Annotation and Visualization ( PAVIS ) analysis [48] . The PAVIS result illustrated the relative distribution of LANA peaks in relation to genes ( Fig 5B ) . A dramatic reduction in LANA peaks were annotated at 5' UTR region in KS tissues as compared to PEL cell lines by 3–5 fold . By analyzing the relative distance from peaks to TSS ( transcription start site ) , we found a completely different distribution pattern of LANA binding peaks at the promoter regions in KS tissues as compared to previous studies in PEL ( Fig 5C ) . By cross-comparing the identified LANA peaks in these two KS tissues , we found very few overlapped peaks as shown in Fig 5D and S1 Dataset . The representative overlapped peak was illustrated and validated in Fig 5E . Further comparing with LANA binding sites in PEL , KSHV infected SLK and endothelial cell lines , we found almost no common sites in KS tissues ( S1 Dataset ) . The difference in LANA binding sites on the host genome in KS tissues might suggest a different role for LANA in KSHV pathogenesis , but could not rule out the possibility that the results arose from the cellular heterogeneity in KS tissues . RTA protein encoded by ORF50 is the key switch regulator that controls KSHV reactivation [13 , 14] . The epigenetic status in RTA region may reflect the state of KSHV infection [21 , 22] . To validate the epigenetic landscape in the classic KS tissues , we carefully examined and verified the histone modifications at the RTA promoter region by ChIP-qPCR . As shown in Fig 6A , the repressive H3K27me3 histone modifications dominated the RTA promoter region ( 69–71 Kb ) in KS tissues of both patients . The results of ChIP-qPCR also confirmed the enrichment of H3K27me3 at the promoter region of RTA and reflected the differences between these two cases ( Fig 6B ) . Moreover , the enrichment of H3K27me3 could be detected at different regions of RTA promoter in the ChIP-qPCR assay ( S2 Fig ) . The GAPDH region exhibited enrichments of AcH3 and little or no H3K27me3 histone modifications , which was the experimental control for the specificity of LANA , AcH3 and H3K27me3 antibodies ( Fig 6B ) . Previous studies in in vitro cell culture systems described a very similar epigenetic landscape at the RTA promoter region as compared to the results in KS tissues [21 , 22] , implicating that the conclusion from in vitro studies about RTA regulation could well support and apply to the in vivo scenario . The TR region of KSHV genome consists of highly repeated sequences of 801 bp with multiple copies which can range over 20 Kb [4] . Previous studies have proved abundant AcH3 histone modifications with LANA accumulation at the TR region in in vitro cell culture systems [21 , 22 , 49] . However , we found that the TR region in classic KS tissues was lacking in AcH3 histone modifications with abundant LANA accumulation ( Figs 3 and 5A ) . To analyze the epigenetic landscape at the TR region without consideration of sequence repetition , we re-aligned the sequence reads for each sample to the TR sequence using Bowtie2 . The reanalyzed epigenetic landscape at the TR region did not change as illustrated in Fig 7A . To verify the results of the ChIP-Seq , we examined the TR region with the same samples by ChIP-qPCR . As shown in Fig 7B , the enrichment of LANA binding and absence of AcH3 histone modifications were confirmed by ChIP-qPCR . To validate the distinct results in KS tissues , we performed the ChIP experiments in Doxycycline inducible recombinant KSHV . 219 harboring SLK ( iSLK . 219 ) and body-cavity-based lymphoma ( BCBL-1 and BC3 ) cell lines using the same protocol . The results were shown in Fig 7C . Very strongly enriched signals of LANA binding and AcH3 histone modifications were observed at the TR region as previously reported . The hyperacetylation of histone H3 at TR region was presumably thought to be involved in the assembly of DNA replication factors , yet the significance remained unknown [49] . TR region contains the latent replication origin of KSHV genome , thus hypoacetylation of histone H3 at the TR region might affect the latent replication of KSHV genome , hampering the maintenance of KSHV episomes [40 , 50 , 51] . This needs further investigation to determine whether losing episomes during the process of in vitro culture of cells derived from KS tissues was related to the absence of AcH3 histone modifications at TR region . The hyperacetylation of histone H3 can introduce a loosened chromatin structure at TR region , which may facilitate a poised chromatin structure at the long unique region for the topological speculation . However , the hypoacetylation of histone H3 at the TR region in classic KS tissues was shown to correlate a silenced state of viral genome , thus the acetylation of histone at the TR region may not be related to KSHV gene expression according to the results . Since KS tumor cells are originated from endothelial cells , we also examined histone modifications and LANA binding sites of KSHV genome in KSHV infected lymphatic endothelial cells ( LEC . KSHV ) . However , the result in LEC . KSHV was also different from the established results in KS tissues . The strong enrichment of LANA binding and AcH3 histone modifications could be observed at the TR region as the same with other in vitro cultured cell lines ( S3 Fig ) . To further confirm the established epigenetic landscape in KS tissues , we examined the epigenetic histone modifications of KSHV genome in new cases of KS tissues ( two classic and one AIDS-related ) . As shown in Fig 8A and 8B , the results of ChIP-qPCR in new cases of classic KS tissue kept good consistency with the previously examined two cases . The established epigenetic landscape at multiple sites were validated , including the TR , RTA promoter , miR-cluster and vIRF3 regions . The general maps of these two cases are illustrated in S4 Fig . Since other subtypes of KS share a common histological characteristic [23] , we wondered whether they would have a similar epigenetic landscape as the classic KS tissues . We additionally determined the epigenetic histone modifications of KSHV genome in one case of AIDS-related KS tissue by ChIP-qPCR . Intriguingly , we found the results in AIDS-related KS tissue are similar to the established one in classic KS tissues , but more enrichment of AcH3 histone modifications were observed at the TR and vIRF3 coding region ( Fig 9 ) . The difference between classic and AIDS-related samples made us speculate that acetylation of histone H3 at the TR and vIRF3 coding region might correspond to the progression of KS disease . The lesions in classic KS cases are generally localized at the extremities with slow or limited progression while the lesions in AIDS-related KS cases usually spread to the whole body and lead to significant mortality [52–56] . The speculation of a relationship between histone acetylation at the TR and vIRF3 coding region and the progression of KS disease will need a larger subset of cases to be examined to support this hypothesis although the data so far is highly suggestive . By analyzing histone modifications and LANA binding sites in classic KS tissues , our study established the epigenetic landscape of KSHV genome in clinical KS samples for the first time . The established epigenetic landscape of KSHV genome in classic KS tissues provided direct evidence to support distinct latent programs in KS tissues . A similar epigenetic landscape was observed at the RTA promoter region in KS tissues as compared to the results from in vitro cell culture systems , which supported the physiological significance of in vitro studies regarding RTA regulation . The distinct AcH3 histone modifications at the TR and vIRF3 coding regions in classic KS tissues provided important clues about the progression of KS disease , which would be a helpful reference for doctors to diagnose clinical patients using epigenetic targeted strategies . We have analyzed the epigenetic landscape of the KSHV genome in classic KS tissues and demonstrated similarities and differences which provide new insights towards understanding KSHV epigenetics , which is important for future studies on the mechanism of KSHV infection and pathogenesis .
Experiments in the present study were conducted according to the principles in the Declaration of Helsinki . The usage of clinical Kaposi's sarcoma ( KS ) tissues was reviewed and ethically approved by the Institutional Ethics Committee of the First Teaching Hospital of Xinjiang Medical University ( Urumqi , Xinjiang , China; Study protocol no . 20082012 ) . Written informed consent was obtained from all participants , and all samples were anonymized . The clinical KS tissues were collected from four patients who had received a pathological diagnosis of KS , including three classic KS and one AIDS-related KS . The patients were of Uygur and Kazak ethnicities from the local region . All samples were collected from Xinjiang province , northwestern China . Details about the patients/specimen were described in the S1 File . Body-cavity-based lymphoma ( BCBL-1 ) cell line was derived from KSHV positive primary effusion lymphoma patients [57] . BCBL-1 and BC3 was maintained in RPMI 1640 medium ( Hyclone ) containing 10% fetal bovine serum ( FBS ) and 5% antibiotics ( penicillin and streptomycin , Hyclone ) . Doxycycline inducible recombinant KSHV . 219 harboring SLK ( iSLK . 219 ) cell line was established by J . Myoung and D . Ganem [58] , and was kindly provided by Fanxiu Zhu ( Florida State University ) . iSLK . 219 cell line was cultured in DMEM ( Hyclone ) supplemented with 10% FBS ( Hyclone ) and 5% antibiotics ( penicillin and streptomycin , Hyclone ) . Lymphatic endothelial cells ( LEC ) were purchased from PromoCell ( C-12216 ) and cultured with Endothelial Cell Growth Medium MV2 kit ( C-22121 , PromoCell ) . The collected fresh clinical KS samples ( 0 . 1–0 . 2 g ) were stored at -80℃ before usage . The protocol of ChIP-Seq in clinical KS samples is described below: ( Optional ) Enhanced Cross-link: Add formaldehyde to a final concentration of 1 . 5% . Incubate at room temperature for 3 min . Quench the cross-linking by adding 0 . 15 mL glycine ( 1 . 25 M ) . Incubate at room temperature for 10 min . Centrifuge at 200 × g for 5 min to remove the supernatants and wash the pellet with PBS twice . *Note the pretreated beads should not be blocked with sperm DNA . Antibodies in the ChIP-Seq experiments: Anti-Trimethyl-histone H3 ( Lys27 ) ( H3K27me3 ) rabbit polyclonal antibody ( 07–449 ) was purchased from Merck Millipore . Anti-acetyl-histone H3 ( AcH3 ) rabbit polyclonal antibody ( 06–599 ) was purchased from Merck Millipore . Anti-LANA mouse monoclonal antibody produced by 1B5 hybridoma was made in our laboratory ( Antigen source for immunization: LANA 900-1162aa ) [45] . The ChIP-Seq data ( data quality parameters were described in the S2 File ) were aligned to human genome ( hg19 ) and KSHV genome ( HQ404500 plus 35 copies of TR [U75699 . 1] ) using Bowtie2 [31]; only one mismatch was allowed . The output files were subjected to peak calling and generation of genome-wide maps with MACS ( Model-based Analysis of ChIP-Seq ) and Homer2 , as described previously [32 , 33] . The Input group was used as control . For the analysis of histone modifications with MACS , the parameters were set according to the protocol as followed:—nomodel , —shiftsize = 73 . The default P value cutoff for the peak detection was 10−5 . . For the analysis of histone modifications with Homer2 , the adjusted parameters were set as followed: -style = histone , -size = 100 or 150 , -minDist = 300 . The default P value cutoff for the peak detection was 10−4 . The final result of identified peaks was generated by the combination of MACS and Homer2 analysis . For the analysis of LANA binding sites with MACS , the parameters were set as previously reported:—nomodel , —shiftsize = 50 . The Input group was used as control . The P value cutoff for the peak detection was 10−3 . Results were visualized by IGV software [59] . Normalization factors were calculated according to the depth of sequencing and formulated as followed: Normalization Factor = Total reads of sample / Total reads of Input . Values on the y axis of each panel in IGV software was adjusted according to the calculated normalization factors ( S3 File ) . The peak information was annotated with Peak Analyzer . The distribution of peaks in relation to genes was calculated by PAVIS [48] . Real-time RT-PCR was performed with a SYBR green Master Mix kit ( Toyobo ) . Reaction mixtures contained 5 μl Master Mix plus Rox , 1 μM each primer , and 4 μl diluted sample . All primers are listed below: The program set on the 7900HT sequence detection system ( Life Technologies ) was 95℃ for 5 min , followed by 40 cycles at 95℃ for 15 s , 58℃ for 20 s and 72℃ for 30 s . Melting curve analysis was performed to verify the specificity of the products and each sample was tested in triplicate . The original data have been submitted to SRA ( Sequence Read Archive ) in NCBI website . The accession number of this project is SRP081036 . KSHV genome: HQ404500 . TR: NC_009333 , 137169–137969 . | Epigenetic modifications are thought to determine the fate of KSHV infection . The epigenetic landscape of KSHV genome in in vitro cell culture systems was well studied previously . However , the physiologically relevant scenario in clinical KS tissues is unclear . In this study , we performed ChIP-Seq experiments in classic KS tissues and mapped out the AcH3 and H3K27me3 histone modifications on KSHV genome , as well as the genome-wide LANA binding sites . The results revealed a similar H3K27me3 landscape but distinct AcH3 patterns on the KSHV genome compared to the results from in vitro cultured PEL and KSHV infected SLK cells . Intriguingly , there were different LANA binding sites seen on the host genome and a reduced number of AcH3 histone modifications at the TR region of KSHV genome were found . The established epigenetic landscape of KSHV genome in classic KS tissues provides new insights towards our understanding of KSHV epigenetics , which is important for future studies on the mechanism of KSHV infection and pathogenesis . | [
"Abstract",
"Introduction",
"Results/Discussion",
"Materials",
"and",
"Methods"
] | [
"medicine",
"and",
"health",
"sciences",
"pathology",
"and",
"laboratory",
"medicine",
"gene",
"regulation",
"pathogens",
"biological",
"cultures",
"cancers",
"and",
"neoplasms",
"microbiology",
"genomic",
"library",
"construction",
"histone",
"modification",
"oncology",
"viruses",
"dna",
"viruses",
"cell",
"cultures",
"dna",
"construction",
"epigenetics",
"dna",
"molecular",
"biology",
"techniques",
"chromatin",
"herpesviruses",
"promoter",
"regions",
"research",
"and",
"analysis",
"methods",
"genomic",
"signal",
"processing",
"chromosome",
"biology",
"medical",
"microbiology",
"gene",
"expression",
"microbial",
"pathogens",
"chromatin",
"modification",
"kaposi's",
"sarcoma-associated",
"herpesvirus",
"kaposi",
"sarcoma",
"sarcomas",
"molecular",
"biology",
"biochemistry",
"signal",
"transduction",
"cell",
"biology",
"nucleic",
"acids",
"dna",
"library",
"construction",
"viral",
"pathogens",
"genetics",
"biology",
"and",
"life",
"sciences",
"cell",
"signaling",
"organisms"
] | 2017 | Epigenetic Landscape of Kaposi's Sarcoma-Associated Herpesvirus Genome in Classic Kaposi's Sarcoma Tissues |
Leptospirosis is a zoonotic disease responsible for high morbidity around the world , especially in tropical and low income countries . Rats are thought to be the main vector of human leptospirosis in urban settings . However , differences between urban and low-income rural communities provide additional insights into the epidemiology of the disease . Our study was conducted in two low-income rural communities near the coast of Ecuador . We detected and characterized infectious leptospira DNA in a wide variety of samples using new real time quantitative PCR assays and amplicon sequencing . We detected infectious leptospira in a high percentage of febrile patients ( 14 . 7% ) . In contrast to previous studies on leptospirosis risk factors , higher positivity was not found in rats ( 3 . 0% ) but rather in cows ( 35 . 8% ) and pigs ( 21 . 1% ) . Six leptospira species were identified ( L . borgpetersenii , L kirschnerii , L santarosai , L . interrogans , L noguchii , and an intermediate species within the L . licerasiae and L . wolffii clade ) and no significant differences in the species of leptospira present in each animal species was detected ( χ2 = 9 . 89 , adj . p-value = 0 . 27 ) . A large portion of the world’s human population lives in low-income , rural communities , however , there is limited information about leptospirosis transmission dynamics in these settings . In these areas , exposure to peridomestic livestock is particularly common and high prevalence of infectious leptospira in cows and pigs suggest that they may be the most important reservoir for human transmission . Genotyping clinical samples show that multiple species of leptospira are involved in human disease . As these genotypes were also detected in samples from a variety of animals , genotype data must be used in conjunction with epidemiological data to provide evidence of transmission and the importance of different potential leptospirosis reservoirs .
Leptospirosis is an infectious disease that causes morbidity in about 1 . 03 million people each year [1] . Severe disease produces multisystem complications such as acute renal or hepatic failure , or severe pulmonary hemorrhage that can lead to death [2 , 3] . Leptospirosis has a widespread distribution , but is mostly prevalent in tropical and poor regions of the world [1] . The causative agents of this zoonotic disease are spirochete bacteria in the genus Leptospira . Bacteria are shed into the environment via urine from infected animals and are transmitted to humans through skin abrasions or mucosal surfaces by direct contact with urine or with contaminated environmental sources [3 , 4] . Rats are thought to play a major role in human infection , and are considered the main reservoir of leptospirosis in urban slums [5 , 6] . High rat density near houses , living or working near garbage , exposure to open sewers , and others risk factors have been identified for leptospirosis [7 , 8] . Likewise , exposure to infected livestock is a recognized risk factor for leptospirosis , but is usually considered an occupational hazard . Leptospirosis infection is also associated with exposure to contaminated environmental sources such as during flooding events , working in rice or cane fields , or when exposed to river water during recreational activities [4 , 9–12] . Leptospirosis in low-income rural areas has been under-reported and under-studied , resulting in large knowledge gaps regarding disease transmission in such places , despite the fact that a large portion of the human population lives in rural areas . Differences in ecology and human behavior between these communities and urban slums , such as animal and pathogen diversity , livestock ownership and recreational activities , might drive differences in the epidemiology of this disease . Here , we present the findings of a two year diversity and prevalence study conducted in rural Ecuador . We collected samples from human febrile patients , cattle , pigs and rats to test for the presence of infectious Leptospira spp . using a newly developed sensitive and specific PCR assay We also genotyped positive samples using an amplicon sequencing approach that allowed us to identify the distribution of genotypes across potential animal reservoirs in an attempt to link host species ( and exclude others ) to human disease .
Our study was carried out from December 2013 to June 2015 , and took place in two rural sites in Manabi Province , Ecuador ( S1 Fig ) . Site 1: Abdon Calderon ( 1° 2' 0" S , 80° 20' 0" W ) , a rural parish near to Portoviejo City . Site 2: the rural communities of Ayacucho , Honorato Vasquez and , La Union Pueblo Nuevo in Santa Ana de Vuelta Larga Parish ( 1°12′25″S 80°22′15″W ) . The distance between Site 1 and Site 2 is approximately 9 Km . The economy of these rural communities is primarily based on agriculture , but subsistence animal ownership is common with animals living in close proximity to houses . Abdon Calderon and Santa Ana de Vuelta Larga were chosen for this study due to the high leptospirosis prevalence recorded by the Ecuadorian Health Ministry in 2012 . Urine samples from cattle and pigs raised in Calderon ( Site 1 ) and Santa Ana ( Site 2 ) were collected from local slaughterhouses . In these communities , animal owners ( or their friends or neighbors ) take their livestock to the slaughterhouses and were able to verify the origin of each animal sample . We visited the slaughterhouses twice a month and the number of samples received ranged from 0 to 10 , depending on the presence of animals from each site ( S1 Table ) . Samples from 48 cattle and 35 pigs were obtained from Site 1 , and 117 cattle and 93 pigs from Site 2 slaughterhouses . Urine was collected directly from bladders of slaughtered animals and transported on ice to our laboratory in Quito , where they were stored at -20°C . Rat snap traps were provided to willing homeowners who were asked to set the traps inside their homes every 15 days . Homeowner’s efforts were likely sporadic and we were notified when trapping was succesfull . Kidneys were obtained from 36 and 65 rats ( Rattus spp . ) from Site 1 and Site 2 respectively , and placed in 90% molecular grade ethanol . DNA was extracted from rat tissue samples within a week of collection . Blood and urine samples were obtained from febrile patients ( 1–6 days of fever ) with no diarrheic or acute respiratory symptoms presenting at Site 1 and Site 2 local Public Health Centers . All samples were collected by the Health Centers for routine diagnostic testing targeting other diseases such as Dengue , Chikungunya , and urinary tract infections . Aliquots of residual samples were stored at -20°C . Throughout a period of 18 months , paired samples ( urine and sera ) from 449 patients were collected at the Site 1 health center . At the Site 2 health center , paired samples were collected from 149 patients while single serum ( n = 72 ) and urine ( n = 10 ) samples were collected from 82 patients . One or 2 drops of fresh human blood were injected in a rubber stopper tube ( for the blood collection ) which contained 7 mls of EMJH semi-solid medium ( supplemented with 200 μg/mL 5- fluorouracil ) and incubated for 1 or 2 weeks at room temperature after which the tubes were transported to the main laboratory where 1ml of the primary culture was inoculated in a new tube containing 7 mls of semi-solid EMJH medium; tubes were incubated for another 4 months at 30°C and monitored for leptospiral growth using a dark-field microscopy . Animal urine was collected directly from bladders at slauterhouses with a syringe , and 1 ml was inoculated into a tube containing 10 mls of liquid EMJH , tubes were inmediatelly transported to the main laboratory ( 10 hours ) at 4°C and subjected to 3 10-fold dilutions in semisolid EMJH medium supplemented with 200 μg/mL 5- fluorouracil . Cultivation was attempted with a total of 56 human blood , 86 pig urine , and 123 cattle urine samples . DNeasy Blood and Tissue kit ( Qiagen , CA , USA ) was used to extract DNA from 200 uL of human sera and from leptospira EMJH isolates . Animal urine and rat kidney DNA was extracted as previously described [13] We searched the 278 publicly available leptospira 16S rRNA gene sequences in the NCBI ( http://www . ncbi . nlm . nih . gov ) and JGI ( https://img . jgi . doe . gov ) databases for phylogenetically informative signatures among Leptospira spp . We used MEGA 6 [14] to align complete 16S rRNA gene sequences and identified single nucleotide polymorphisms ( SNP ) that provided discriminatory power among and within “pathogenic”and “intermediate” clades . Such SNP signatures were identified in a 153 bp region of the gene . This region corresponds to positions 3102729 to 3102577 and 1935913 to 1936065 in Leptospira interrogans serovar Lai str . 56609 ( AE010300 ) . For other pathogen species , TaqMan MGB assays have been used to detect very low amounts of target DNA and discriminate among species and even strains [15–18] and were therefore used here . In order to create positive controls and standardize quantification , a 330 bp fragment of the 16S rRNA gene from a “pathogenic” species ( Leptospira interrogans Lai ) , and an “intermediate” species ( Leptospira licerasiae VAR010 ) were synthesized as gBlocks gene fragments ( IDT ) and inserted inside the pCR 2 . 1 TOPO vector ( Invitrogen Corp . , Carlsbad , CA , USA ) . These fragments included the 153 bp fragment that provides identification and discrimination of “pathogenic” and “intermediate” leptospira ( Table 1 ) . We designed two single probe assays: ( i ) assay “111” detects all “pathogenic” and “intermediate” , but not “saprophytic” leptospira; ( ii ) assay “50” detects only “pathogenic” leptospira . TaqMan MGB probes and primers ( Table 1 ) were designed using Primer Express Software ( Life Technologies ) . Assays were designed as single probe assays that amplify the same region but probes anneal to different targets . Both TaqMan MGB single-probe assays were run using a 7900HT Fast Real-Time PCR System ( Applied Biosystems ) with SDS v2 . 4 software . A total reaction volume of 10 μl was prepared by using 1x TaqMan Genotyping Master Mix ( Applied Biosystems by Life Technologies , Foster City , CA , USA ) , 1μM of each primer , 300nM of each probe ( Table 1 ) , and 1μl of DNA . Thermal cycling conditions for the two assays ( 111 and 50 ) were as follows: 50°C for 2 min . , 95°C for 10 min . , followed by 45 cycles of 95°C for 15 sec . , 58°C for 1 min . Assay metrics were determined by testing their performance across several parameters: accuracy , specificity , limit of quantification ( LoQ ) and detection ( LoD ) , and linearity , as described in Price et al . [15] . SNP signatures used for probe design were subjected to both in silico ( BLAST analysis ) and laboratory screening to determine their accuracy towards leptospira species ( S2 Table ) ; all probes were specific to the target Leptospira clade . DNA from 15 leptospira species used in this study was supplied by the Royal Tropical Institute , Amsterdam , the Netherlands ( L . alexanderi , L . borgpetersenii , L . interrogans , L . kirschneri , L . kmetyi , L . noguchii , L . santarosai , L . weilii , L . fainei , L . inadai , L . licerasiae , L . wolffii , L . biflexa , L . vanthielii , L . wolbachii ) . Non- leptospira species were tested to determine accuracy of assays towards leptospira species . These species were Acinetobacter baumanii , Klebsiella pneumoniae , Staphylococcus epidermidis , Escherichia coli , Enterococcus faecalis , Enterobacter aerogenes , Moraxella catarrhalis , Streptococcus agalactiae , Neisseria meningitides , and Listeria monocytogenes ( S2 Table ) , and were chosen because they belong to genera that are either common pathogens or commonly associated with humans and other animals . For laboratory testing , the Leptonema illini 16S rRNA complete gene ( chosen because it is the nearest neighbor of Leptospira genus ) was synthetized by gBlocks gene fragments ( IDT ) and inserted inside the pCR 2 . 1 TOPO vector ( Invitrogen Corp . , Carlsbad , CA , USA ) to keep it stable . Known quantities of control vectors ( see positive control construction ) were used for determination of lowest LoQ ( S2 Table ) and LoD ( S3 Table ) . The lowest LoQ was defined when 4 of 4 replicates amplified with a cycle threshold ( CT ) of <0 . 3 standard deviation from the mean CT . The lower LoD was measured , after defining the lowest LoQ , as the lowest concentration of analyte that gave rise to signal ( considering that negative controls gave no signal ) . Range of linearity of each assay was determined by 10-fold dilutions that resulted in a Ct separation of about 3 . 4 . Loss of linearity was defined as the lowest dilution point where this separation was seen . We also tested robustness of the assays by varying the annealing temperature from 56°C to 60°C . All samples were tested with assays 111 and 50 , additionally and in order to test PCR inhibitor compounds that may lead to false-negative results , we amplified a fragment of the beta-actin gene [19] in ten percent of human and animal samples , this fragment is expected to be present in all pig , rat , cattle and human tissue . No PCR inhibition was found . Amplicons of all positive samples for assays 111 and 50 were diluted to 10^5 , and re-amplified using the same forward and reverse primers ( Table 1 ) but containing a universal tail for indexing [20] . Briefly , primers for specific amplicons were designed in order to be compatible with Illumina adapter sequences , amplicons were reamplified in a conventional thermocycler with the following conditions: 94°C for 5 min . , followed by 15 cycles of 94°C for 30 sec . , 60°C for 30 sec . , 72°C for 2 min . , and a final extension at 72°C for 5 min . Amplicons were pooled and sequenced with the Illumina MiSeq in order to confirm the results of the assays and identify leptospira species present in each sample . Raw sequences were demultiplexed with the MiSeq software and also with the fastq-multx tool from ea-util package [21] . Consensus sequences were obtained using DNASTAR Seqman software for sequence assembly and contig management . For pathogenic ( and intermediate ) species , sequences matched a single species with 100% identity , providing confidence in our species assignments ( S2 Fig ) . When samples were available , serum samples from patients positive for PCR ( Leptospira DNA in urine or sera ) were tested for IgM antibodies using the Panbio Leptospira IgM ELISA ( Queensland , Australia ) . MAT ( Microscopic Aglutination Test ) was also performed on these samples at the national reference laboratory: Instituto Nacional de Salud Pública e Investigación , Guayaquil-Ecuador . Phylogenetic reconstruction of amplicon sequencing results was carried out in MEGA6 [14] using the Maximum Likelihood method , with the Kimura 2-parameter model [22] . This model had the lowest Bayesian Information Criterion score and corrected Akaike information criterion as determined through model selection analysis in MEGA . Cluster confidence was determined by performing 100 bootstrap replicates . Association analysis of leptospirosis cases ( n = 227 ) and rainfall from Site 1 ( Calderon-Manabi ) was performed using data from the INHAMI ( National Institute of Meteorology and Hidrology ) data-base ( http://www . inamhi . gob . ec/ ) and serologically confirmed leptospirosis cases provided by the Ecuadorian Health Ministry ( MSP-VGVS-2015-0197-O ) . Only data from Site 1 was used as meteorological data from Site 2 were not available . For this , we defined high and low precipitation thresholds using local rainfall data; we defined a month with low precipitation as accumulated rainfall of ≤50 mm and a month of high rainfall with precipitation >50 mm . In order to account for the time between infection and clinical onset of the disease , monthly leptospirosis cases were matched with precipitation values from the same and previous months . Association between variables was performed in the SPSS program by running a Generalized Linear Equation model using the Wald statistic and the Maximum likelihood estimate . Significant differences among all pairs were identified using Fisher's exact test . Homogeneity of proportions was tested with post hoc pair-wise comparisons [23] . P-values were adjusted with the Benjamini & Hochberg method [24] , and the odds ratio probability was calculated by using the Epitools package in R [25 , 26] . Human samples were collected by Ecuadorian Health Ministry Technicians and analyzed with the authorization of The Bioethical Committee of Universidad San Francisco de Quito ( study code 2011–40 ) and by the Northern Arizona University Institutional Review Board ( 482212–1 ) . Cattle and pig urine samples were collected from animals slaughtered for human consumption at the local abattoir and were thus processed as normal work of the abattoir . Verbal consent from the animal owners was provided . Rattus spp . were collected by homeowners using methods consistent with AVMA guidelines for euthanasia and approved by the NAU IACUC ( Protocol 13–006 ) .
Real-time PCR assays described here can detect infectious leptospira species ( excluding “saprophytic” species ) and can discriminate among members of the “pathogenic” and “intermediate” clades . As predicted by in silico analysis , assays 111 and 50 were 100% specific for species within the “pathogenic” and “intermediate” clade , and for the “pathogenic” species , respectively . Both assays 111 and 50 must be used to define species in the “intermediate” clade; assay 50 is needed to exclude “pathogenic” species from those detected using Assay 111 . None of the saprophytic leptospira species , the 10 non- leptospira species , or the closely related Leptonema illini amplified with either assay ( S1 Table ) . Both assays exhibited the lowest limit of quantification ( LoQ ) and lowest limit of detection ( LoD ) to be one 16S rRNA copy per microliter of extracted DNA ( S3 Table ) . Range of linearity of assays 111 and 50 was 10^8 to10^1 16S rRNA genes . Assay 111 is robust along a 4°C variation in annealing temperature ( 56–60°C ) and assay 50 provided robust amplification along a 2°C variation in annealing temperature ( 56–58°C ) . Amplicon sequencing also enabled separation of the main “pathogenic” and “intermediate” clades but provided topological resolution within these clades similar to that provided by others who used a 1 , 230 bp fragment ( Levett et al . 2015 ) to describe the molecular phylogenetics of the Leptospira genus ( S2 Fig ) . Sequencing this 153 bp region allowed us to discriminate among most of the species of leptospira found in animal urine , rat kidney , and human urine and sera samples . The sequences obtained matched 16S sequences from known species at 99–100% ( S6 Table ) . From Site 1 , samples from febrile patients with no diarrhea or acute respiratory symptoms were positive for pathogenic leptospira DNA in 17 . 3% ( n = 78/449 ) of the samples: sera 2% , ( n = 8/449 ) , and urine 15 . 6% ( n = 70/449 ) . Site 2 patients showed positivity in 9 . 5% ( n = 22/231 ) : 3 . 6% in sera ( n = 8/219 ) , and 8 . 8% in urine ( n = 14/159 ) . We never detected Leptospira DNA in sera and urine from the same patient . Urine from slaughtered cattle and pigs , and rat kidney were collected in both sites over the same time period when human samples were obtained ( Fig 1 ) . Positivity in cattle urine was 35 . 4% ( n = 17/48 ) and 35 . 9% ( n = 42/117 ) for Site 1 and Site 2 , respectively . Lower positivity was found in pig urine: 5 . 7% ( n = 2/35 ) for Site 1 and 26 . 9% ( n = 25/93 ) Site 2 . Rat kidney positivity was low for both sites: 2 . 8% ( n = 1/36 ) in Site 1 , and 3 . 1% ( n = 2/65 ) in Site 2 ( Table 2 ) . Additionally , we recovered leptospira isolates from 1 out of 56 ( L . santarosai ) human blood samples and 2 ( L . interrogans ) out of 123 cattle urine; drafts of these genomes were deposited in GenBank [27] . Six different leptospira species were identified in both study sites: L . borgpetersenii , L kirschnerii , L santarosai , L . interrogans , L noguchii , and an intermediate species within the L . licerasiae and L . wolffii clade ( Fig 1 and S4 Table ) . Our amplicon sequencing data divide L . interrogans into two genotypes . Despite Site 1 being 9 km away from Site 2 , the frequency of leptospira species circulating within each site did not differ between sites ( χ-squared = 9 . 89 , adj . p-value = 0 . 27 ) or between 2014 and 2015 ( χ-squared = 5 . 16 , adj . p-value = 0 . 56 ) . Additionally , there were no significant differences in the species of leptospira present in cattle and pigs ( χ-squared = 10 . 06 , adj . p-value = 0 . 124 ) . We also found that L . interrogans and L . borgpetersenii are more likely to be carried by cattle ( β > 1 . 08 , p-value < 0 . 005 , β > 1 . 25 , p-value < 0 . 005 ) . We also characterized the serological diversity on some febrile patient samples that were collected one to six days after the onset of fever . Thirteen of 16 DNA-positive sera samples and 74 of 84 DNA-negative sera samples from patients who had DNA-positive urine were tested with a commercial leptospira IgM ELISA and with MAT by the local reference laboratory . ELISAs of all patients with DNA-positive sera samples ( representing acute infections ) were negative and only one was MAT positive to serovar Canicola . As such , we were not able to determine the rate of false negative detection of PCR methods . Results on DNA-negative sera samples from febrile patients with DNA-positive urine also showed low positivity; only 3 were positive for ELISA and 20 for MAT . Association analysis using leptospirosis IgM positivity recorded by the Health Ministry for 2011–2014 from Site 1 and monthly rainfall showed that 30 . 9% of positive sample variation could be explained by rainfall in the same month ( R22011-2014 = 0 . 309 , CI: 0 . 027–0 . 545 , p-value = 0 . 032 ) . In contrast , 57% of variation in positive cases could be explained by rainfall from the preceding month ( R22011-2014 = 0 . 57 , CI:0 . 36–0 . 75 , p-value = 1 . 18 x 10−5 ) . For each year , the percentage of cases explained by rainfall in the same month ranged from 12–61% ( R22011 = 0 . 56 , CI: -007-0 . 8 , p-value = 0 . 05; R22012 = 0 . 12 , CI: -0 . 4–0 . 64 , p-value = 0 . 7; R22013 = 0 . 51 , CI: -0 . 095–0 . 83 , p-value = 0 . 090; R22014 = 0 . 61 , CI:0 . 024–0 . 9 , p-value = 0 . 044 ) . In contrast with the percentage of cases explained by rainfall in the preceding month that ranged from 51–69% ( R22011 = 0 . 67 , CI: 0 . 11–0 . 9 , p-value = 0 . 024; R22012 = 0 . 58 , CI: 0 . 012–0 . 86 , p-value = 0 . 047; R22013 = 0 . 51 , CI: -0 . 095–0 . 83 , p-value = 0 . 093; R22014 = 0 . 69 , CI:0 . 199–0 . 9 , p-value = 0 . 0123 ) . Analysis of the presence of Leptospira DNA in urine and serum samples from all febrile patients who presented to the local health center between 2014 and the beginning of 2015 showed that the odds of a febrile patient being infected with Leptospira was not higher in months with high rainfall ( ORSite 1 = 0 . 95 , CI: 0 . 55–1 . 6 , p-value = 0 . 85;ORSite 2 = 1 . 5 , CI: 0 . 6–3 . 9 , p-value = 0 . 4 ) . Similar results were obtained when the number of cases per month was matched with precipitation from the previous month ( Table 3 ) .
Very little is known about leptospirosis in Ecuador or the epidemiology of leptospirosis in low-income , rural communities . The aim of our study was to better understand some of the factors thought to influence occurrence and transmission of pathogenic leptospira in these areas . These factors include animal reservoirs and rainfall . Detection of pathogenic leptospira is difficult due to the practical complications of culturing , the diversity of hosts , genetic diversity of the pathogen , and heterogeneous or absent symptoms [3 , 28 , 29] . Molecular detection of pathogenic leptospira is , in general , more sensitive than culture or the microscopic agglutination tests ( MAT ) [28 , 30] . However , currently used molecular detection assays suffer from suboptimal sensitivity and inability to discriminate between species in the “pathogenic” and “intermediate” clade [31–37] . Recently , species in the “intermediate” clade have been associated with disease [37–40] , presenting the need to include them in epidemiological studies . Moreover , “intermediate” leptospira have been shown to infect animals and humans in Ecuador [13] . The methodology presented here allowed us to detect and discriminate between DNA from “pathogenic” and “intermediate” leptospira species in clinical and animal samples . In contrast with previous reports of rats being the main reservoirs of leptospirosis in urban slums , [5 , 6 , 41–43] , our results show low positivity in rats but high positivity in cattle ( Table 2 ) . This suggest that in our study sites and during our sampling period , cattle are likely to be more important reservoirs for leptospirosis than rats . Leptospirosis is commonly associated with rainfall [9 , 44] . Regression analysis from serological data on human cases of leptospirosis collected by the Health Ministry for 2011–2014 showed a strong association with rainfall . Interestingly , our analyses of Leptospira DNA-positive results from sera and urine samples collected at this same site between 2014 and the beginning of 2015 show no such association . Our 16 DNA-positive sera samples , however were collected in the rainy season ( n = 15 ) or in the following month ( n = 1 ) . In contrast , the 84 DNA-positive urine samples show no association with rainfall , possibly due to our inability to attribute infection to a specific month . Detection of Leptospira DNA in urine and not serum is likely due to sampling after the bacteria have been cleared from the bloodstream . Leptospira is not typically detected in serum after 5 days of fever , but can be found in urine months later [4] . Antibodies can be detectable by MAT before and after bacteria are cleared from circulation , thus paired samples showing a temporal increase in titer indicate acute cases [4] . Of the 84 DNA-positive urine samples , MAT analyses were conducted on 74 and 20 ( 27% ) were MAT positive . As convalescent samples are never obtained by the Health Ministry , these results do not allow us to discriminate between acute and long-term disease . Conversely , 54 of these samples ( 73% ) were MAT negative , indicating chronic carriage ( assuming that present serovars can be adequately detected with the serological test ) . As chronic leptospirosis is usually asymptomatic [45 , 46] , we suspect that the fever of these patients and presentation at the clinic might have been due to another illness . Overall , our results suggest that rainy seasons are important for acute infections , but also suggest that chronic carriage is common , the implications of which deserve further exploration . Previous work on leptospirosis epidemiology suggests that some serotypes are more likely to infect certain hosts [29 , 47 , 48] , therefore the genetic diversity in the population of hosts may also be species specific . Thus , finding the same species in both rats and humans [39] may provide evidence for an epidemiological link . In contrast , our genotyping work shows that different host species are routinely infected by a variety of circulating leptospira species and these leptospira were not differentially associated with certain host species . Importantly , all sampled animal types contained leptospira species that were also associated with human clinical samples , similar to what has been reported in Southeast Asia [36] . This suggests that genotype associations ( at a species level of resolution ) alone cannot be used to exclude any of these animal types as potential contributors to human transmission . Higher resolution genotyping methods and epidemiological information such as evidence of direct contact with a given animal would be needed in addition to genotype information to establish an epidemiological link . Given the high prevalence in cattle and their shared genotypes with humans , we hypothesize that cattle are playing an important roll in the transmission to humans . However other animals also share genotypes with humans . Distribution and morbidity of leptospirosis depends on geographic and socio-economic features [1] . Therefore differences between urban and rural leptospirosis should be expected . Our study provides strong evidence that suggest that leptospirosis in low-income rural communities might be different than in urban slums . People living in low-income rural areas usually share their peri-domestic and recreational environment with livestock and in some cases with sylvatic animals . This contrasts with urban areas where contact with these animals is mostly encountered in occupational settings . Our study also exhibits the importance of considering other animals besides rats as important reservoirs of the disease . It also suggests that some rural areas might have higher diversity of pathogenic leptospira species than what has been shown for urban-slums . We believe that it is important to understand leptospirosis ecology in low-income rural areas , especially considering the high burden of leptospirosis in rural areas [1] , and that in Latin America , 30–60% of people live in rural areas [49] . | Leptospirosis is a febrile disease responsible for high morbidity rates all around the world . It is caused by spirochete bacteria in the genus Leptospira and transmitted to humans through animal urine or contaminated soil or water . The epidemiology of leptospirosis has been extensively studied in urban communities , where rats are thought to be the main animal reservoir of the disease . However , leptospirosis has been under-reported and under-studied in low-income rural areas where different conditions may result in different transmission risks . In two low-income rural communities near the coast of Ecuador , we detected and characterized leptospira from febrile patients and putative animal reservoirs . Our results show the complexity of leptospirosis epidemiology and suggest that livestock may serve as an under recognized source of leptospirosis . Understanding the epidemiology of leptospirosis in many settings provide insights into transmission that can ultimately be used to to prevent and control this disease . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] | [
"livestock",
"medicine",
"and",
"health",
"sciences",
"leptospira",
"body",
"fluids",
"pathology",
"and",
"laboratory",
"medicine",
"ruminants",
"pathogens",
"tropical",
"diseases",
"microbiology",
"vertebrates",
"animals",
"mammals",
"urine",
"bacterial",
"diseases",
"neglected",
"tropical",
"diseases",
"bacteria",
"bacterial",
"pathogens",
"veterinary",
"science",
"infectious",
"diseases",
"swine",
"veterinary",
"diseases",
"zoonoses",
"medical",
"microbiology",
"microbial",
"pathogens",
"leptospirosis",
"agriculture",
"anatomy",
"physiology",
"biology",
"and",
"life",
"sciences",
"leptospira",
"interrogans",
"cattle",
"amniotes",
"bovines",
"organisms"
] | 2016 | High Leptospira Diversity in Animals and Humans Complicates the Search for Common Reservoirs of Human Disease in Rural Ecuador |
Highly repetitive and transposable element rich regions of the genome must be stabilized by the presence of heterochromatin . A direct role for RNA interference in the establishment of heterochromatin has been demonstrated in fission yeast . In metazoans , which possess multiple RNA–silencing pathways that are both functionally distinct and spatially restricted , whether RNA silencing contributes directly to heterochromatin formation is not clear . Previous studies in Drosophila melanogaster have suggested the involvement of both the AGO2-dependent endogenous small interfering RNA ( endo-siRNA ) as well as Piwi-interacting RNA ( piRNA ) silencing pathways . In order to determine if these Argonaute genes are required for heterochromatin formation , we utilized transcriptional reporters and chromatin immunoprecipitation of the critical factor Heterochromatin Protein 1 ( HP1 ) to monitor the heterochromatic state of piRNA clusters , which generate both endo-siRNAs and the bulk of piRNAs . Surprisingly , we find that mutation of AGO2 or piwi increases silencing at piRNA clusters corresponding to an increase of HP1 association . Furthermore , loss of piRNA production from a single piRNA cluster results in genome-wide redistribution of HP1 and reduction of silencing at a distant heterochromatic site , suggesting indirect effects on HP1 recruitment . Taken together , these results indicate that heterochromatin forms independently of endo-siRNA and piRNA pathways .
In D . melanogaster , an estimated one-third of the genome is composed of repetitive and noncoding sequences associated with a condensed form of chromatin known as heterochromatin . Heterochromatin is characterized by repeat-rich sequences , hypoacetylation of histone tails , and dimethylation of histone H3 on lysine 9 ( H3K9me2 ) [1] . A conserved nonhistone Heterochromatin Protein 1 ( HP1 ) is a critical component of heterochromatin , localizing predominantly at and near centromeres but also residing at telomeres , and the Y and fourth chromosomes . These regions tend to be rich in transposable elements ( TEs ) , which must be suppressed in order to maintain genomic stability but can serve a cellular function , particularly in the case of Het-A and TART at the telomeres ( reviewed in [2] ) . The phenomenon of position-effect variegation ( PEV ) provided the first glimpse into the role of heterochromatin in gene silencing in Drosophila . When a normally euchromatic gene is relocated near heterochromatin , variegated expression results from variable levels of heterochromatin spreading over the gene in each cell . Screens for dominant mutations that either suppress {Suppressor of variegation [Su ( var ) ]} or enhance {Enhancer of variegation [E ( var ) ]} PEV were performed to identify key components of heterochromatin . For example , mutation of Su ( var ) 3-9 , which encodes an H3K9 methyltransferase , was identified in a large screen for modifiers of PEV [3] . Accordingly , loss of HP1 , encoded by Su ( var ) 2-5 , causes increased expression of a gene subject to PEV while an extra copy has the reverse effect [4] . Pioneering genetic and biochemical studies in Schizosaccharomyces pombe have shed considerable light on mechanisms of heterochromatin assembly . The RNA interference ( RNAi ) machinery was found to play a key role in heterochromatin formation by detecting the transcription of specific DNA repeats located at the mating type locus and the centromere and subsequently nucleating heterochromatin . For example , double-stranded RNAs ( dsRNA ) produced by bidirectional transcription of pericentromeric repeats are processed by the RNase III endonuclease Dicer1 into short interfering RNAs ( siRNAs ) [5] . The Argonaute1 PAZ and PIWI domain protein binds these siRNAs as part of the RNA-induced transcriptional silencing complex ( RITS ) [6] . Loading of RITS with siRNA and recruitment of the complex to the site of dsRNA transcription requires the Clr4 histone methyltransferase , which methylates H3K9 [7] . This methylation mark serves as a binding site for Swi6 , a fission yeast homolog of HP1 , leading to heterochromatin establishment and spreading . Importantly , heterochromatin can also be nucleated independently of RNAi by other mechanisms . For example , in the absence of RNAi the ATF/CREB stress-activated proteins promote heterochromatin formation at the mating type locus [8] , and the Taz1 protein can establish HP1 recruitment to telomeres [9] . These studies exemplify the redundancy of RNAi and additional mechanisms with respect to the formation of heterochromatin . All RNA silencing pathways are characterized by the activity of an Argonaute effector protein that binds directly to small RNA . The five Argonautes in Drosophila can be divided into two families based on homology . The AGO subfamily includes AGO1 and AGO2 , and the Piwi subfamily consists of Piwi , Aubergine ( Aub ) , and AGO3 ( reviewed in [10] ) . AGO1 and AGO2 are expressed throughout the fly while piwi , aub , and AGO3 are expressed mainly , although not exclusively , in the gonad [11]–[13] . AGO1 is required for the microRNA pathway , which regulates mRNA expression and functions chiefly through translational repression . Protecting against exogenous double stranded RNA , AGO2 associates with 21–22 nt siRNA produced by Dicer-2 ( Dcr-2 ) , and this pathway is required for viral immunity and a robust RNAi response [14] , [15] . In addition , AGO2 also binds endogenous siRNAs ( endo-siRNAs ) , the majority of which silence the expression of TEs outside of the gonad [16]–[19] . Suppression of TEs is especially imperative in the gonad in order to limit the propagation of unwanted mutations and is achieved principally by the activity of the Piwi subfamily proteins . Piwi , Aub , and AGO3 bind to 23–30 nt RNAs termed Piwi-interacting RNAs ( piRNAs ) that are predominantly derived from genomic locations termed piRNA clusters [20] , [21] . These piRNA producing loci are mainly pericentromeric and enriched in transposon sequences . From these and previous studies , it became clear that the piRNA pathway exists to eliminate TE transcripts in the gonad [22]–[24] . Based on comparative sequence analysis of piRNAs immunopurified from the ovary , the “ping-pong” or “amplification loop” model for germline piRNA biogenesis was proposed [20]–[23] . Precursor transcripts from piRNA clusters , derived from either one or both strands [25] , give rise to piRNAs bound by Piwi , Aub , or AGO3 . Those piRNAs antisense to a homologous TE transcript can result in its cleavage , and this event defines the 5′ end of a secondary piRNA that can then bind and cleave an antisense piRNA cluster transcript , and the cycle can continue . Piwi appears to play a minor role in ping-pong piRNA amplification [25] , [26] , which is thought to occur primarily in the cytoplasmic nuage where Aub and AGO3 localize [20] , [23] , [27] . In contrast , Piwi resides in the nucleus [28] . Production of precursor transcripts at certain piRNA clusters that give rise to piRNAs from both sense and antisense strands ( dual-strand clusters ) is dependent on the germline specific HP1 homolog Rhino [29] . Rhino functions specifically in the ping-pong pathway , acting upstream of Aub and AGO3 but not Piwi . Piwi independently serves an additional role in the silencing of certain TEs expressed in somatic follicle cells surrounding the ovary . This somatic piRNA pathway depends on Piwi alone and therefore does not undergo ping-pong amplification [25] , [26] , [30] . The flamenco ( flam ) piRNA cluster , which controls the gypsy , ZAM , and Idefix retrotransposons [31] , [32] , is one of the major sites of primary piRNA production [25] , [26] , [30] , [33] . Piwi associates with piRNAs generated by flam and other piRNA clusters and has been proposed to cleave homologous TE transcripts using its Slicer activity [22] . Previous studies suggest that one or more RNA silencing pathways may participate in transcriptional TE silencing by inducing heterochromatin formation . First , mutation of AGO2 results in pleiotropic cellular defects in early embryos including mislocalization of HP1 and the histone H3 variant CID , which binds specifically the centromere [34] . Later in development , AGO2 mutants display mislocalization of HP1 on polytene chromosomes of the larval salivary gland [35] . Additionally , silencing of a pericentromeric transcriptional reporter is relieved when the maternally derived pool of AGO2 is reduced . Despite these defects , AGO2 mutant flies develop normally and are fertile , suggesting that these defects are mild and can be compensated by other mechanisms . Several pieces of evidence implicate piRNA pathways in establishment or maintenance of heterochromatin in the soma . First , mutation of piwi , aub , or spn-E , encoding an RNA helicase required for the germline piRNA pathway [24] , [26] , results in defects in heterochromatic silencing and visible changes in heterochromatin localization . These mutants reduce silencing of pericentromeric transcriptional reporters and exhibit mislocalization of HP1 and H3K9me2 in salivary gland polytene chromosomes [36] . Moreover , a recent study identified HP1 as an interactor of Piwi in yeast two-hybrid screens [11] . The two proteins coimmunoprecipitate from embryonic nuclear lysate and display partially overlapping localization patterns in polytene chromosomes . Furthermore , both proteins associate specifically with the chromatin of two transposable elements , 1360 and the F element . Based on their findings , the authors propose that Piwi could serve as a recruitment platform for HP1 binding . This model appears not to be applicable to the 3R-TAS subtelomeric region , a site of Piwi chromatin association and piRNA production [21] . Mutation of piwi results in an increase of HP1 association and an increase of transcriptional silencing at 3R-TAS . It remains an open question whether other sites in the genome could serve as Piwi-dependent HP1 recruitment sites . In other metazoans , it is similarly unclear whether RNA silencing can establish heterochromatin directly . A recent study in chicken indicates that a 16 kb constitutive heterochromatin domain that separates the folate receptor gene and the β-globin locus is maintained by a Dicer and Argonaute 2 ( cAgo2 ) dependent mechanism [37] . Intriguingly , cAgo2 was shown to associate with the heterochromatic domain by chromatin immunoprecipitation ( ChIP ) suggesting a direct effect . However , it is not known whether this represents a general mechanism to maintain heterochromatin . In this study , we investigated whether HP1 association with heterochromatin in Drosophila is mediated by either the AGO2 dependent endo-siRNA pathway or by piwi dependent piRNA pathways . Using transcriptional reporters and ChIP , we show that piRNA clusters are subject to heterochromatic silencing and bound by HP1 . Interestingly , mutation of AGO2 , piwi or aub results in increased silencing at piRNA clusters and an increase in HP1 association with these loci . Furthermore , loss of piRNA production at a single piRNA locus results in global redistribution of HP1 and a reduction of silencing at a distant heterochromatic site . Therefore , our results indicate that HP1 can associate with chromatin independently of both endo-siRNA and piRNA pathways .
We sought to determine if HP1 is recruited to heterochromatin by AGO2 or Piwi . The majority of genomic regions that produce the bulk of piRNA , termed piRNA clusters , are pericentromeric and rich in transposable elements [20] , [21] . These regions also produce endo-siRNA [16]–[19] , and due to their proximity to the centromere , may be heterochromatic and serve as platforms for Argonaute mediated HP1 recruitment . In order to test genetically whether pericentromeric piRNA clusters are heterochromatic , we examined a collection of fly lines bearing P element transgene insertions inside or in close proximity to four piRNA producing loci , flam , 80EF , 42AB , and 38C . The P elements contain a mini-white transcriptional reporter that was assayed for expression in the adult eye . Genomic locations of these transgene insertions are indicated in relation to previously identified small RNAs immunoprecipitated with Piwi , Aub/AGO3 , and AGO2 respectively from various cell types ( Figure 1 , Figure S1 ) [16] , [17] , [20] , [21] . Lines harboring P elements inside or in the vicinity of a piRNA cluster exhibit variegating coloration of distinct eye facets similar to PEV , suggesting the presence of variably spreading heterochromatin at their sites of insertion ( Figure 2 , Table 1 ) . Interestingly , insertions within a piRNA cluster that display high mini-white expression without variegation harbor SUPor-P constructs , which contain Suppressor of Hairy wing [Su ( Hw ) ] insulator sequences that flank and likely protect the mini-white reporter from the effects of surrounding heterochromatin [38] . Expression analysis of these transcriptional reporter insertions indicates that piRNA clusters and their immediate vicinity are subject to HP1 dependent silencing . Reporter expression levels of three lines harboring an insertion at flam , 80EF , or 42AB with the most apparent variegation were tested for dependence on heterochromatin . P{EPgy2}DIP1EY02625 is inserted in a gene located on the centromere distal side of the flam piRNA producing locus on the X chromosome ( Figure 1A ) , PBac{PB}c06482 resides within the 80EF cluster on chromosome 3L ( Figure 1B ) , and P{EPgy2}EY08366 borders the centromere proximal edge of the 42AB piRNA locus on chromosome 2R ( Figure 1C ) . In order to test whether these reporters are sensitive to perturbation of heterochromatin , the expression of mini-white was examined in Su ( var ) 2-505/+ and Su ( var ) 3-91/+ dominant loss-of-function mutants , which are compromised for HP1 and H3K9 methyltransferase activity respectively . As expected , decreased silencing of mini-white expression resulting in increased pigmentation was observed for all three insertions in the heterochromatin mutants compared to wild type ( Figure 2 ) , suggesting that the vicinity of P element insertion are indeed heterochromatic . We next tested whether the transcriptional reporters at piRNA clusters are sensitive to perturbations in the piRNA and endo-siRNA silencing pathways . If Piwi were responsible for direct recruitment of HP1 to piRNA clusters , mutation of piwi should increase mini-white expression similarly to disruption of heterochromatin . Surprisingly , piwi1/piwi2 loss-of-function mutants exhibit a substantial loss of reporter expression indicating increased silencing when compared to wild type ( Figure 2 ) . Furthermore , aubQC42/aubΔP-3a loss-of-function piRNA pathway mutants result in a similar reduction of mini-white expression . Strikingly , the flam transcriptional reporter expression level was decreased dramatically in the transheterozygous endo-siRNA pathway mutant , AGO251B/AGO2414 compared to wild type ( Figure 2A ) . Similarly , in the AGO251B null mutant , the 42AB transcriptional reporter displays almost complete silencing ( Figure 2C ) . Spectroscopic analysis of extracted eye pigment verifies the overall changes in mini-white expression levels for each genotype compared to wild type ( Figure 2D ) . Additionally , examination of Dcr-2L811fsX mutants shows a similar mild increase in silencing for the transcriptional reporter inserted near flam ( Figure S2A ) . The opposite effects of piRNA and endo-siRNA pathway mutations compared to heterochromatin mutations suggest that these RNA silencing pathways may actually oppose heterochromatin formation at piRNA clusters . In order to further examine the heterochromatic nature of piRNA clusters at higher resolution , ChIP assays were performed in adult heads to assess HP1 association with two piRNA clusters , flam and 80EF , in the soma . Genomic locations of primer sets that uniquely amplify regions spanning these piRNA clusters are indicated in Figure 1A and 1B . As positive controls , primers for two transposable elements known to recruit HP1 , TART , a telomere-specific non-LTR retrotransposon , and 1360 , a DNA transposon were also tested [39]–[40] . Euchromatic genes hsp26 and yellow were also included in the analysis as negative controls for HP1 association . In wild type fly heads , HP1 is observed at or near locations that give rise to piRNAs and endo-siRNAs at both flam and 80EF loci . ChIP was performed using α-HP1 antibodies in chromatin prepared from wild type heads , and the amount of DNA associated was determined by quantitative PCR using specific primer sets . As expected , low levels of hsp26 and yellow are immunoprecipitated with HP1 , while TART and 1360 levels are enriched above the euchromatic genes by over six-fold ( Figure 3 ) . At flam , HP1 associates with the majority of regions that produce high levels of piRNAs or endo-siRNAs approximately two to three-fold over the euchromatic sites ( Figure 3A , primer sets 1–15 ) . Similarly , at 80EF , HP1 immunoprecipitates piRNA and endo-siRNA producing regions two to three-fold higher than the negative controls indicating the presence of heterochromatic marks at these loci ( Figure 3B , primer sets G-M ) . Regions flanking these areas display approximately one to two-fold enrichment over euchromatic sites , which may be due to tapering of HP1 spreading ( Figure 3B , primer sets A-F and N-P ) . ChIP using antibodies directed against the chromatin insulator protein Su ( Hw ) verified its presence at known insulator sequences gypsy and 1A-2 [41] but only background levels at TART , 1360 , and piRNA clusters , indicating the specificity of HP1 association at these sites ( Figure S3 ) . Rabbit IgG negative control immunoprecipitations yielded negligible amounts of DNA for all sites tested ( <0 . 3% input ) . Consistent with the transcriptional reporter assay , RNA silencing mutants display elevated levels of HP1 at piRNA clusters . ChIP of HP1 was performed in piwi1/piwi2 mutant heads , and similar levels at positive and negative controls were obtained compared to wild type ( Figure 3 ) . In contrast , at the flam locus , a two to five-fold increase in HP1 levels is observed at the centromere proximal side of the locus compared to wild type ( Figure 3A , primer sets 6–15 ) . Little change in HP1 recruitment is observed at the centromere distal end of flam in piwi1/piwi2 mutants ( Figure 3A , primer sets 1–5 ) . At 80EF , HP1 levels increase two to three-fold in piwi1/piwi2 mutants compared to wild type across all primer sets examined ( Figure 3B , primer sets A-P ) . In order to address differences in strain background and potential accumulation of TEs in piwi mutant strains , we performed ChIP assays comparing piwi1/piwi2 mutants to a piwi1/+ heterozygous strain and obtained similar results ( Figure S4 ) . ChIP experiments performed in AGO251B mutant heads show a similar overall increase of HP1 at piRNA clusters compared to piwi1/piwi2 mutants . Levels of HP1 at hsp26 , yellow , TART , and 1360 are similar in AGO251B mutants and wild type while differences are apparent at piRNA clusters ( Figure 3 ) . At flam , AGO251B mutants display a two to seven-fold increase of HP1 association with the centromere proximal side compared to wild type ( Figure 3A , primer sets 6–15 ) . At the centromere distal end , no significant changes in HP1 levels are detected ( Figure 3A , primer sets 1–5 ) . For 80EF , AGO251B mutants show similar levels of HP1 to wild type at the centromere distal end ( Figure 3B , primer sets A-D ) while an approximately two to five-fold increase of HP1 is detected in the remainder of the regions tested ( Figure 3B , primer sets E-P ) . Moreover , ChIP assays in AGO251B homozygous mutants compared to an AGO251B/+ heterozygous strain produced similar results ( Figure S5 ) . Similar to AGO251B mutants , Dcr-2L811fsX mutants show an increase of HP1 at regions that produce small RNAs compared to wild type ( Figure S2B and S2C ) . HP1 protein levels in wild type , piwi1/piwi2 , and AGO251B fly heads are similar indicating that the increased chromatin association observed is not due to an increased amount of HP1 ( Figure S6 ) . The increased HP1 chromatin association with piRNA clusters in RNA silencing mutants compared to wild type is consistent with increased silencing of P element insertions , and these results suggest that at least some of the observed effects on reporter gene expression in RNA silencing mutants are due to chromatin related events . Taken together , these data suggest an antagonistic effect of Piwi , Aub , and AGO2 on HP1 recruitment to chromatin in somatic tissue . Given the evidence that transposable elements are mainly silenced in the gonad via piRNA pathways and in the soma via the endo-siRNA pathway , we wanted to determine whether HP1 also associates with piRNA clusters in gonadal tissues . Therefore , we investigated HP1 recruitment to piRNA clusters in wild type ovaries by ChIP . As in heads , low levels of hsp26 and yellow are immunoprecipitated with HP1 , whereas TART and 1360 levels are enriched above the euchromatic genes by over ten-fold ( Figure 4 ) . At the flam locus , a four to fifteen-fold increase over the euchromatic sites in HP1 levels is observed at most sites at the centromere proximal side of the locus ( Figure 4A , primer sets 4–15 ) . Similarly , at 80EF , HP1 immunoprecipitates small RNA producing regions two to twenty-fold higher than euchromatic sites indicating the presence of heterochromatic marks at these loci ( Figure 4B , primer sets A-P ) . Rabbit IgG negative control immunoprecipitations yielded negligible amounts of DNA for all sites tested . We were unable to immunoprecipitate DNA at levels above background from either heads or whole ovaries using multiple antibodies to Piwi , Aub , AGO3 , and AGO2 that have been used in previous studies for immunoprecipitation or immunofluorescence ( data not shown ) [11] , [22] , [23] , [42] . We wished to address whether HP1 association with piRNA clusters is dependent on Piwi in the gonad , which express high levels of both proteins . Due to a complete loss of germ cells and the severe underdevelopment of ovary tissue in piwi mutants , it was not possible to obtain enough mutant material to perform ChIP . Therefore , we examined the recruitment of HP1 to chromatin in an ovarian somatic follicle cell line ( OSC ) that expresses Piwi but not Aub or AGO3 and produces only primary piRNAs , a large proportion of which derive from the flam locus [30] . The majority of Piwi was depleted from OSC cells by siRNA-mediated knockdown , and depletion of Piwi does not affect HP1 or Lamin protein levels compared to mock transfected cells ( Figure 5A ) . Subsequently , we investigated HP1 recruitment to piRNA clusters by ChIP in OSC cells . In mock treated cells , low levels of hsp26 and yellow are immunoprecipitated with HP1 , while TART and 1360 levels are enriched above the euchromatic genes by 1 . 5- to over two-fold ( Figure 5B and 5C ) . Two additional TEs tested , gypsy and mdg1 , are immunoprecipitated at similar levels to TART with HP1 ( Figure 5B–5E ) . At flam , HP1 associates with the piRNA cluster similar to TE levels ( Figure 5B and 5C ) . Despite much lower piRNA production from the 80EF cluster in OSC compared to flam [30] , HP1 associates with piRNA producing regions of 80EF at similar levels to flam and TEs ( Figure 5C , primer sets A-P ) . Overall , the HP1 recruitment profile in OSC is similar to that of heads and whole ovaries albeit at lower relative levels . In Piwi knockdown cells , no significant differences are seen for HP1 recruitment to all sites compared to mock treated cells except a two-fold decrease at the 1360 element . Rabbit IgG negative control immunoprecipitations yielded low amounts of DNA for all sites tested ( <0 . 06% and <0 . 07% input for mock and Piwi knockdown cells , respectively ) . Importantly , Piwi association with chromatin is detectable in OSC cells , but its profile differs from that of HP1 . In mock treated cells , antibodies directed against Piwi [22] immunoprecipitate euchromatic sites at levels similar to that of TEs ( Figure 5D and 5E ) . Furthermore , the majority of regions producing piRNA at flam is also immunoprecipitated at comparable levels to both euchromatic sites and TEs ( Figure 5D ) . Moreover , levels of Piwi association with 80EF is akin to that of flam , while several sites in both flam and 80EF clusters show particular enrichment of Piwi up to three-fold compared to the average association with other sites tested ( Figure 5D and 5E ) . In Piwi knockdown cells , Piwi chromatin association drops two to five-fold , down to background levels at all sites except for some residual association with two sites in or near the flam locus . Mouse IgG negative control immunoprecipitations yielded low amounts of DNA in comparison to α-Piwi immunoprecipitations in mock treated cells for all sites tested ( <0 . 04% and <0 . 02% input for mock and Piwi knockdown cells , respectively ) . We conclude that in ovarian somatic follicle cells , reduction of the total pool of Piwi as well as the chromatin bound fraction does not affect HP1 association with piRNA clusters and has a minimal effect on HP1 association with TE chromatin association . We next sought to determine whether loss of piRNA production at a single piRNA cluster would affect HP1 recruitment to chromatin . Previous studies have shown that mutation of various RNA silencing components results in global mislocalization of HP1 on polytene chromosomes [35]–[36] . Mutation of flam has been previously shown to result in loss of piRNA production [20] and upregulation of the gypsy retroelement [32] . In order to obtain a genome-wide view of HP1 chromatin association in flam mutants , we examined the localization of HP1 to highly replicated salivary gland polytene chromosomes from either wild type or flam1 mutant third instar larvae by indirect immunofluorescence using α-HP1 antibodies . In wild type , HP1 localizes predominantly to a concentration of heterochromatin where the centromeres of each chromosome coalesce , termed the chromocenter ( Figure 6A , green ) . In contrast , flam1 mutants display expansion of HP1 at the chromocenter . Spreading of HP1 is apparent on the second and third chromosomes , but not on the X chromosome , where flam is located . As a reference , we also examined the localization of the chromatin insulator protein Mod ( mdg4 ) 2 . 2 , which is unchanged in localization between wild type and flam1 ( Figure 6A , red ) . The extent of HP1 chromocenter expansion is comparable to the level of HP1 expansion that we observe in spn-EhlsE1/spn-EhlsE616 mutants ( Figure S7 ) . A lesser degree of HP1 expansion was also observed in flamBG02658/flamKG00476 mutants ( data not shown ) . Finally , total HP1 levels are unchanged in flam1 whole flies compared to wild type ( Figure S6 ) . These results indicate a global change in HP1 localization resulting from inactivation of a single piRNA cluster . We reasoned that accumulation of HP1 at the chromocenter of flam1 mutants may result in an increase in silencing at pericentromeric sites . Therefore , the expression of transcriptional reporters at 42AB or 80EF piRNA clusters , which are located on different chromosomes from the flam locus , was examined in flam1 mutants . Compared to wild type , flam1 mutants harboring a P element insertion at either 42AB or 80EF piRNA clusters display mildly decreased pigmentation suggesting increased silencing at these distinct pericentromeric loci ( Figure 6B ) . Finally , to verify HP1 genome-wide redistribution in flam1mutants , we examined the effect of flam1 on the silencing of a centromere distal heterochromatic site on a different chromosome . The DX1 transgene array consists of seven mini-white P elements with one inverted copy at a normally euchromatic site at 50C on chromosome 2R [43] . Due to this configuration , the array forms ectopic repeat induced heterochromatin and displays a variegated phenotype similar to PEV that is dependent on HP1 . Expression of the DX1 array was assessed based on variegation of eye pigmentation in wild type , heterozygous flam1/+ , and homozygous or hemizygous flam1 mutants ( Figure 6C ) . Due to a wide range of eye coloration , variegation was scored by categorization into five groups that ranged between Light ( few pigmented facets ) to Dark ( almost all facets pigmented ) . For females , 3% of wild type was classified as Dark , while 29% of flam1/+ and 52% of flam1 mutants displayed the same high level of pigmentation . In males , 15% of wild type was scored as Medium-Dark or Dark while 40% of flam1 males fell into these categories . These results indicate that mutation of flam can suppress heterochromatic silencing in trans . Taken together with the HP1 centromeric expansion in polytene chromosomes and increased pericentromeric silencing in flam1 mutants , there appears to be a global redistribution of HP1 resulting from the loss of piRNA production from a single locus .
Several reasons dictated the choice of piRNA clusters as the focus of our analyses . First , both endo-siRNAs and piRNAs are generated from these loci [16]–[21] . Next , we reasoned that at least some piRNA clusters are likely to be heterochromatic because of their strong bias toward TE-rich pericentromeric positions in the genome [20] , [21] , in close proximity to the vast majority of HP1 localization . In fact , early cloning attempts determined that the flam locus is located in a repetitive , TE rich heterochromatic region [44] . Furthermore , the pericentromeric position of these clusters likely coincides with the transition between euchromatin and heterochromatin , corresponding to the borders of HP1 spreading . This characteristic allows variegation assays , which monitor the variable spreading of HP1 and heterochromatin , to be extremely sensitive . ChIP assays at the borders of HP1 spreading would also likely be optimally sensitive to both local and overall changes in HP1 chromatin association . Finally , piRNA clusters contain enough unique sequence for specific primer design and monitoring by directed ChIP analysis . Given that AGO2 is the predominant Argonaute expressed outside the gonad that participates in the silencing of TEs in the soma , we tested whether AGO2 could recruit HP1 to chromatin in somatic tissue . Moreover , it has been shown that AGO2 mutants exhibit mislocalization of HP1 [34] , [35] . However , our results show that mutation of AGO2 results in a strong increase of silencing of transcriptional reporters at or near piRNA clusters and a mild increase of HP1 chromatin association in heads . Given the extent of increased silencing in the AGO2 mutant compared to piwi or aub mutants , which accumulate HP1 on chromatin to a similar degree , a posttranscriptional step of silencing likely contributes to the negative effects observed on transcriptional reporters . AGO2 mutants show a plethora of cellular defects during early nuclear divisions but develop normally and are fertile suggesting that effects on these various processes as well as HP1 localization are mild or otherwise compensated [34] . Therefore , AGO2 is unlikely to be required for HP1 recruitment in this tissue . Additionally , we find that HP1 association at piRNA clusters does not depend on the presence of Piwi . Our analysis of piRNA clusters included flam , a primary piRNA cluster , and 80EF , a germline piRNA producing locus . We examined both flam and 80EF clusters in somatic head tissue and ovaries , which are a mixed population of somatic follicle and germline derived cells . In heads , there is no apparent requirement for piwi with respect to HP1 recruitment to the piRNA clusters or to TEs that were examined . In our study , Piwi chromatin association was detected only in OSC cells , and its presence is dispensable for HP1 chromatin association . The flam piRNA cluster produces high levels of primary piRNA in OSC while 80EF is active for piRNA production in germ cells but not in OSC [25] , [26] , [30] . Nonetheless , Piwi associates with both the flam and 80EF clusters at comparable levels , suggesting that the amount of piRNA production from a particular locus does not correlate with Piwi chromatin association . Furthermore , the pattern of Piwi chromatin association in OSC differs from that of HP1 in that there is no particular enrichment of Piwi at TEs above euchromatic sites and only a minor accumulation at a few sites in the flam and 80EF piRNA clusters . When Piwi levels were reduced by siRNA knockdown , Piwi chromatin association was essentially abolished but HP1 recruitment was not affected except for a two-fold decrease over the 1360 element . Previous studies suggested that the 1360 element may be responsible for nucleating heterochromatin on the largely heterochromatic fourth chromosome and further showed that mutation of factors representing all RNA silencing pathways , piwi , aub , spn-E , Dcr-1 , and Dcr-2 , affect 1360 dependent heterochromatic silencing [40] , [45] . Unlike the results in adult heads , no accumulation of HP1 over piRNA clusters was detected as a result of Piwi knockdown in OSC cells . This discrepancy may reflect differential effects in distinct cell types or the length of the Piwi knockdown in OSC cells , which was at least adequate to essentially eliminate Piwi chromatin association . In a related but independently derived ovarian somatic follicle cell line ( OSS ) , Piwi and HP1 do not colocalize in the nucleus [33] , and this finding supports the conclusion that Piwi does not direct HP1 recruitment in this cell type . Also consistent with our results , HP1 remains localized to the chromocenter in salivary gland polytene chromosomes in piwi null mutants [11] , [36] . We conclude that association of HP1 with chromatin can occur independently of AGO2 and piwi in somatic tissue . A previous study addressed the role of the germline piRNA pathway in HP1 association with transposable elements . The spn-E gene controls predominantly germline piRNA production but does not affect the somatic piRNA pathway [26] . ChIP was used to show that spn-E mutants display significantly decreased levels of H3K9me3 and HP1 at telomeric Het-A but similar to wild type HP1 levels at the I-element and copia TEs , which are distributed throughout the genome [46] . This modest reduction of HP1 at Het-A was apparent in ovaries but not in carcasses , which contain only somatic tissue . One caveat to this study is that ChIP was performed using primers that detect all TEs matching a particular sequence , thus measuring average HP1 and H3K9me levels on TEs across the genome . Nonetheless , this work suggests a limited role for the germline piRNA pathway in HP1 recruitment at the telomere . Several studies have shown that Piwi associates with at least some heterochromatic sites in the genome , but direct evidence that any of these sites serve as recruitment platforms for HP1 and subsequent spreading is lacking . The best characterized Piwi-associated site is the heterochromatic 3R-TAS subtelomeric region , which generates the abundant Piwi bound 20nt 3R-TAS piRNA . Surprisingly , the role of piwi at this location is transcriptional activation , as piwi mutants display increased transcriptional silencing of a nearby reporter transgene as well as an increase of HP1 association at 3R-TAS [21] . Likewise , we observe a mild corresponding increase in HP1 association and silencing at piRNA clusters in piwi mutants suggesting that piwi function could in fact oppose HP1 recruitment at multiple sites in the genome . Our results are consistent with the possibility that piRNA clusters act as boundaries to the spread of pericentromeric heterochromatin . The mechanism of Piwi dependent transcriptional activation has not been determined , but considering that Piwi interacts with the chromoshadow domain of HP1 [11] , Piwi may compete for binding with other HP1 interactors such as Su ( var ) 3–9 that promote heterochromatic silencing . The majority of Piwi protein is found in both somatic and germline tissues of the gonad , yet piwi clearly exerts an effect on non-gonadal somatic tissues as well . RT-PCR analysis shows that piwi transcript is readily detectable outside the gonad and in somatic cell lines [11] , [12] , but Piwi protein is difficult to detect [11] . Nevertheless , mutation of piwi suggests important functions for this gene outside of the gonad . For example , piwi is essential for viability , and loss-of-function mutants display a variety of phenotypes manifest in various non-gonadal somatic tissues such as demonstrated in this study and others , which show a requirement for piwi in pairing-dependent silencing , nucleolar integrity , and chromatin insulator function [47]–[50] . For each of these chromatin related studies , it remains a possibility that even a small amount of maternally deposited Piwi could trigger early events in the oocyte or embryo that persist throughout development , manifesting phenotypes visible in adult somatic tissues . Our results along with previous studies have demonstrated that HP1 mislocalizes from the chromocenter in a subset of piRNA pathway mutants . We found that polytene chromosomes of flam1 mutants exhibit expanded HP1 chromocenter distribution . This result is intriguing because the flam1 mutation affects a single piRNA cluster on the X chromosome but HP1 spreading to other chromosomes is apparent . A previous study detected spreading of HP1 to euchromatic arms especially in spn-E mutants [36] , and we confirmed this result albeit to a lesser degree , with spreading being comparable to the extent seen in flam1 mutants . Perhaps the increase of TE expression in RNA silencing mutants can stimulate HP1 recruitment and spreading from the centromere , which contains the highest concentration of TEs . In fact , transcription of pericentromeric repeats stimulates RNAi-dependent heterochromatin formation in fission yeast [51]–[53] . Redistribution of HP1 in RNA silencing mutants may indirectly affect silencing at various heterochromatic locations in the genome . Seemingly inconsistent with HP1 spreading , spn-E , aub , and piwi mutants display decreased silencing of P element transgene arrays such as DX1 and single insertions at pericentromeric regions on chromosomes 2 and 4 [36] . In our study , we found that mutation of flam also results in loss of silencing at DX1 , which is distant from the flam locus . This reduced silencing in trans could not be due to posttranscriptional events as there are no shared sequences between DX1 and the flam locus . Therefore , we consider the possibility that there exists a finite pool of HP1 that accumulates at the centromere in flam and other RNA silencing mutants at the cost of reduced density and reduced silencing at other heterochromatic regions such as the transgene array , the fourth chromosome , and the telomere . The concept of a limited population of HP1 was suggested previously to explain the finding that the Y chromosome behaves as a suppressor of variegation by acting as a sink for HP1 [43] . Studies in multiple organisms have identified or suggested alternative mechanisms to RNA silencing for the recruitment of HP1 to chromatin . In fission yeast , overlapping and redundant RNAi-dependent and independent mechanisms of heterochromatin formation have been elucidated . In mouse cells , HP1 localization to pericentromeric heterochromatin was found to be RNase A sensitive suggesting that an RNA moiety may be involved in HP1 recruitment [54] . Our data indicate that heterochromatin can form independently of RNA silencing in Drosophila . It will be interesting to determine if any of these alternative mechanisms of heterochromatin formation are conserved throughout evolution .
Fly stocks were maintained at 25°C on standard cornmeal medium . Lines containing P{EPgy2}DIP1EY02625 and P{EPgy2}EY08366 were obtained from the Bloomington Drosophila Stock Center , and a line harboring PBac{PB}c06482 was obtained from the Exelixis Collection at Harvard Medical School . Genomic coordinates of these P-element insertions were confirmed by PCR with primers specific to the P-elements and flanking genomic sequences followed by sequencing . For transcriptional reporter assays , transgenes were crossed or recombined into mutant backgrounds and scored against crosses to yw67c23 as a reference . For ChIP and immunofluorescence , Oregon-R was used as a wild type control . The y v f mal flam1/FM3 stock was selected for heterozygous females each generation to prevent mobilization and accumulation of TEs . For the DX1 variegation assay , DX1/CyO was crossed to y w v f mal flam1/FM7c; CyO/Sp flies or yw67c23; CyO/Sp as a reference . Eye pigmentation of 40 to 60 adult males six days of age was examined , and representative eye photos were taken . To quantify overall levels of eye pigmentation , the heads of 25 male flies of each genotype were dissected , and eye pigmentation was measured as previously described [36] . Briefly , heads were homogenized in 0 . 8 ml of methanol , acidified with 0 . 1% HCl and centrifuged . The absorbance of the supernatant was measured at 480 nm . Adult fly heads or ovaries were dissected and crosslinked with 1 . 8% formaldehyde for 20 min at 23°C . Chromatin was fragmented to an average size of 300 bp by sonication and incubated with antibodies overnight at 4°C . Quantitative PCR was conducted on Applied Biosystems Real Time PCR system using SYBR Green incorporation ( Affymetrix/USB ) . Amplicon sizes ranged between 150 and 250 bp . Chromatin was immunoprecipitated with the following antibodies: α-HP1 ( Covance ) , α-Su ( Hw ) , α-Piwi ( P3G11 , a gift from M . Siomi ) , and normal rabbit or mouse IgG ( Santa Cruz Biotechnology ) . A recombinant N-terminal His-tagged fusion protein of the N-terminal of Su ( Hw ) ( amino acids 1–218 , kind gift of M . Labrador ) was purified from E . coli on a nickel-agarose column and used to immunize guinea pigs using standard procedures . Similar results were obtained using the C1A9 α-HP1 antibody ( Developmental Studies Hybridoma Bank ) , but lower quantities of DNA were obtained . Fifty to one hundred fly heads and twenty five to fifty ovaries were used per IP . The quantities of target genomic regions precipitated by different antibodies were calculated as percent input based on four-point standard curves constructed from input DNA for each primer set . Standard deviation of each PCR performed in quadruplicate was calculated to determine the error of measurement . Two independent ChIP samples were analyzed , and similar results were obtained . ChIP primers were designed to be unique , detecting only sequences present in the flam and 80EF piRNA loci and verified by in silico PCR . All primers ( Table S1 and Table S2 ) were checked for both specificity and efficiency by standard agarose gel electrophoresis and real time PCR respectively . Primers to piRNA clusters amplify in the same DNA dilution range as primers specific to hsp26 and yellow single copy genes compared to high copy TE elements ( Figure S8 ) . Primers to the flam locus were verified to amplify approximately two-fold more DNA from female compared to male genomic DNA . A detailed version of this protocol is available in Text S1 . The OSC line was maintained and Piwi siRNA knockdown was performed as previously described [30] . Briefly , 3×106 trypsinized cells were resuspended in 0 . 1 mL of Solution V of the Cell Line Nucleofector Kit V ( Amaxa Biosystems ) and mixed with 200 pmol of siRNA duplex . Transfection was conducted according to the manufacturer's protocol using the nucleofector program T-029 , and the transfected cells were incubated at 25°C for 48 hrs . Protein knockdowns were verified by Western blotting , and ChIP assays were performed on mock and piwi siRNA transfected cells ( 5×106 cells per IP ) . Preparation and immunostaining of salivary gland polytene chromosomes was performed as described previously [55] . Primary antibodies directed against HP1 ( Covance ) and Mod ( mdg4 ) 2 . 2 ( generated similarly as in [56] ) and Alexa Fluor 488 labeled anti-guinea pig or Alexa Fluor 594 labeled anti-rabbit secondary antibodies ( Invitrogen-Molecular Probes ) were used . The chromosomes were viewed using a Leica epifluorescence microscope and photographed using a Hamamatsu digital camera . Eye pigmentation of 100 to 200 flies was scored . The scoring of variegation was categorized into five groups: Light , Medium-Light , Medium , Medium-Dark and Dark corresponding to the percentage of pigmented facets . Percentage of flies falling into each category was graphed . Representative eye photos were taken . | One role for silent heterochromatin is to preserve the integrity of the genome by stabilizing regions rich in repetitive sequence and mobile elements . Compaction of repetitive sequences by heterochromatin is needed to prevent genome rearrangement and loss of genetic material . Furthermore , uncontrolled movement of mobile elements throughout the genome can result in deleterious mutations . In fission yeast , one important mechanism of heterochromatin establishment occurs through RNA interference , an RNA–dependent gene silencing process . However , it is unclear whether a direct role for RNA silencing in heterochromatin formation is conserved throughout evolution . In the fruit fly , Drosophila melanogaster , which harbors multiple RNA–silencing pathways that are both functionally distinct and spatially restricted , previous studies have suggested the involvement of the endogenous small interfering RNA ( endo-siRNA ) and Piwi-interacting RNA ( piRNA ) pathways in heterochromatin formation . These small RNA silencing pathways suppress the expression of mobile elements in the soma or in both somatic and germline tissues , respectively . Utilizing complementary genetic and biochemical approaches , we monitored the heterochromatin state at discrete genomic locations from which both types of these small RNAs originate in endo-siRNA or piRNA pathway mutants . Our results indicate that heterochromatin can form independently of these two small RNA silencing pathways . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"molecular",
"biology/chromatin",
"structure",
"molecular",
"biology/rna-protein",
"interactions",
"computational",
"biology/transcriptional",
"regulation",
"genetics",
"and",
"genomics/gene",
"expression",
"biochemistry/transcription",
"and",
"translation",
"genetics",
"and",
"genomics/epigenetics",
"developmental",
"biology/developmental",
"molecular",
"mechanisms",
"cell",
"biology/gene",
"expression"
] | 2010 | HP1 Recruitment in the Absence of Argonaute Proteins in Drosophila |
The C-terminal region of the minor structural protein of potato leafroll virus ( PLRV ) , known as the readthrough protein ( RTP ) , is involved in efficient virus movement , tissue tropism and symptom development . Analysis of numerous C-terminal deletions identified a five-amino acid motif that is required for RTP function . A PLRV mutant expressing RTP with these five amino acids deleted ( Δ5aa-RTP ) was compromised in systemic infection and symptom expression . Although the Δ5aa-RTP mutant was able to move long distance , limited infection foci were observed in systemically infected leaves suggesting that these five amino acids regulate virus phloem loading in the inoculated leaves and/or unloading into the systemically infected tissues . The 5aa deletion did not alter the efficiency of RTP translation , nor impair RTP self-interaction or its interaction with P17 , the virus movement protein . However , the deletion did alter the subcellular localization of RTP . When co-expressed with a PLRV infectious clone , a GFP tagged wild-type RTP was localized to discontinuous punctate spots along the cell periphery and was associated with plasmodesmata , although localization was dependent upon the developmental stage of the plant tissue . In contrast , the Δ5aa-RTP-GFP aggregated in the cytoplasm . Structural modeling indicated that the 5aa deletion would be expected to perturb an α-helix motif . Two of 30 plants infected with Δ5aa-RTP developed a wild-type virus infection phenotype ten weeks post-inoculation . Analysis of the virus population in these plants by deep sequencing identified a duplication of sequences adjacent to the deletion that were predicted to restore the α-helix motif . The subcellular distribution of the RTP is regulated by the 5-aa motif which is under strong selection pressure and in turn contributes to the efficient long distance movement of the virus and the induction of systemic symptoms .
Potato leafroll virus ( PLRV ) is the type member of the Poleroviruses , in the family Luteoviridae . These viruses and the related Luteoviruses , collectively referred to here as luteovirids , are phloem limited viruses and transmitted by aphids in a circulative , nonpropagative manner [1 , 2] . Intact virions are required to move long distance in plant hosts and aphid vectors [1 , 3 , 4] . In addition to the coat protein ( CP ) that is required for virion assembly and virus movement , there are three additional virus proteins , the minor capsid protein known as the readthrough protein ( RTP ) , the P17 movement protein , and the P3a protein that play a role in virus movement in the plant [5–11] . P17 localizes to plasmodesmata at the companion cell-sieve element boundary and facilitates the cell-to-cell movement of assembled virus particles in a host-dependent manner [5 , 8] . A small non-AUG-initiated ORF encodes the P3a protein that is required for virus long-distance movement [9] . The RTP , encoded by ORF 3 and 5 , contains the 23 kDa CP and the 57 kDa read-through domain ( RTD ) [12] . The full-length RTP can be detected readily in infected plant tissues , but in purified virus preparations a significant portion of the C-terminal half of the RTD is proteolytically processed yielding a 51–58 kDa RTP [12 , 13] . The RTD has a highly conserved N-terminal region and a variable C-terminal region . The N-terminus of the RTD is important for mediating RTP incorporation into the virion [12] , and is required for aphid transmission and interaction with aphid endosymbiont proteins [14–16] . The variable C-terminal region is dispensable for aphid transmission , but it does play a role in tissue tropism , viral movement , virus accumulation , and symptom development in plant hosts [17–20] . Involvement of the full-length luteovirid RTP in virus movement was previously demonstrated by monitoring in planta progression of viral mutants either unable to synthesize the RTP or bearing small deletions in the C-terminal domain [10 , 18 , 19] . Rodriguez-Medina et al [20] reported that a C-terminal RTD deletion mutant of turnip yellows virus ( TuYV-ΔRTCter ) affected long-distance trafficking in a host-specific manner . In addition , while replacement of portions of the C-terminal RTD by GFP did not prevent the systemic invasion of non-inoculated leaves , virus titer was reduced and no symptoms were observed [21 , 22] . This further supports the hypothesis that the C-terminal half of the RTD contains domains responsible for viral efficient infection and symptom development . The variable C-terminal region of the luteovirid RTD was determined to be highly disordered and involved in protein-protein interactions [23] , but the nature of these interactions is not well studied . Several studies have reported that mutants with altered RTD function are under strong selection to acquire additional mutations that restore normal translation and function of the RTP [10 , 11 , 16] , allowing the virus to move efficiently in phloem and accumulate to wide-type levels in plants . Poleroviruses have monopartite , linear , single-strand RNA ( + ) genomes of 5 . 3–5 . 7 kb in size and they generate subgenomic RNAs ( sgRNA ) for expression of 3’ proximal genes [12 , 24] . Three sgRNAs ( sgRNA1 , sgRNA2 and sgRNA3 ) have been described for PLRV [25] . Poleroviruses are transmitted by aphids directly into phloem tissue [1] . The virus replicates most efficiently in phloem companion cells and then moves through plasmodesmata between companion cells or into sieve elements for long distance transport [16 , 26] . Plant viruses move intracellularly using the cellular cytoskeleton and/or endomembrane system to the cell periplasm and plasmodesmata [27–29] . PLRV requires assembled virions to move systemically in plants , but it is unknown if local cell-to-cell movement requires intact particles [30] . The understanding of viral movement between sieve elements and companion cells is limited , but in general virions require additional viral proteins to facilitate systemic infection of plants [31 , 32] with a few notable exceptions [33–35] . Pore–plasmodesma units are specialized plasmodesmata that have branched connections with the companion cells and small pore-like openings into sieve elements [36–38] . Although the size exclusion limits of pore-plasmodesmata are greater than simple plasmodesmata connecting other cell types ( e . g . mesophyll cells ) , they would prevent the free movement of virions and ribonucleic complexes [30 , 34 , 39] . Cucumber mosaic virus ( CMV ) infection can enable phloem loading and long-distance trafficking of GFP [40] , and the CMV 3a movement protein fused to GFP does traffic through pore-plasmodesmata [41] . Tobacco mosaic virus ( TMV ) can also enhance its access to the phloem of mature plant tissues through the targeted disruption of auxin/indole acetic acid ( Aux/IAA ) transcriptional regulators that control expression of host genes involved in virus cell-to-cell movement , plasmodesmata gating [42] . These findings provide experimental evidence that virus infection can alter the size exclusion limit for plasmodesmata interconnecting the sieve element–companion cell complex . Once viruses enter the phloem , they are passively carried with the flow of photoassimilates and unload in sink tissues , apparently without the same level of control that accompanied their entry [43 , 44] . Here we further investigated the functional role of the C-terminal domain of the RTP in planta and identified an important conserved aromatic amino residue in a conserved α-helix structural motif . These structural features , located near the C-terminus of the RTD of all PLRV isolates , are critical for RTP function . Naturally occurring pseudo-reversions of mutations that disrupt this domain restored protein structure and functions associated with phloem loading .
The C-terminal half of the PLRV RTD has been shown to be important for virus systemic movement efficiency and symptom development [10 , 26] , but the specific motifs directing these activities are unknown . A series of C-terminal RTD deletion mutants were constructed in an infectious PLRV clone ( Fig 1A ) and inoculated into the natural host Solanum sarrachoides ( hairy nightshade , HNS ) . The ability of the mutants to systemically infect and induce symptoms in HNS was quantified by DAS-ELISA ( Fig 1B ) and monitored by observation of interveinal chlorosis symptoms five weeks post inoculation ( wpi ) ( S1 Fig ) . All PLRV C-terminal mutants were infectious and were able to systemically infect plants ( Fig 1B and S1A Fig ) . However , mutants that included deletions of nucleotides 5671–5685 ( e . g . Mut-Δ5670 , Mut-Δ5670–5685 ) accumulated to low levels in systemically infected tissues and typical interveinal chlorosis symptoms were not observed . Deletions downstream of nucleotide 5685 ( Mut-Δ5685 and Mut-Δ5700 ) behaved like wild-type ( WT ) PLRV ( Fig 1B and S1A and S1B Fig ) . Taken together , these data defined nucleotides 5671–5685 as an essential motif responsible for efficient systemic virus accumulation and symptom development in HNS . To determine whether the nucleotides or the encoded five amino acids ( aa ) were important for virus function , a stop codon ( TAA ) was inserted after nucleotide 5670 ( Mut-5670-TAA ) . The low virus accumulation of Mut-5670-TAA and a lack of symptoms suggested that the five amino acids ( leucine , phenylalanine , glutamic acid , tyrosine and glutamine , LFEYQ ) , not the encoding nucleotides are important for virus function ( Fig 1B and S1 Fig ) . The role and function of some polerovirus proteins in systemic movement and virus accumulation are often host specific [8 , 20 , 45] . WT-PLRV , Mut-Δ5670 , Mut-Δ5685 , and Mut-Δ5670–5685 were individually infiltrated into the experimental host Nicotiana benthamiana , as well as Physalis floridana ( groundcherry ) and potato ( S . tuberosum cultivar ‘Red Maria’ ) . Virus accumulation was measured by DAS-ELISA in infiltrated leaves at 3 days post infiltration ( dpi ) ( Fig 2 ) and in young systemically infected leaves at 5 wpi ( Fig 1C and 1D ) . Virions , viral RNAs and RTP all accumulated to similar levels in the leaves infiltrated with each of the mutants ( Fig 2 ) , however , systemic infection and symptom development varied in the different hosts ( Fig 1C and 1D and S1 Fig ) . All the mutants and WT-PLRV accumulated in systemically infected tissues of N . benthamiana and P . floridana , but similar to HNS , Mut-Δ5670 and Mut-Δ5670–5685 , lacking the 5-aa motif , accumulated to significantly lower levels than WT-PLRV or Mut-Δ5685 ( Fig 1C and 1D and S1A Fig ) . Mut-Δ5670 , Mut-5670-TAA and Mut-Δ5670–5685 did not systemically infect potato ( S1A Fig ) . The relative differences between the titers of either Mut-Δ5670 or Mut-Δ5670–5685 and WT-PLRV were less in N . benthamiana , and P . floridana than in HNS . Nevertheless , symptom expression was delayed until 6–7 wpi; two weeks after symptom expression in plants infected with WT-PLRV and the virus titer in N . benthamiana , and P . floridana plants infected with Mut-Δ5670 and Mut-Δ5670–5685 remained significantly lower than in WT-PLRV or Mut-Δ5685 infected plants ( S1C and S1D Fig ) . The mutants all retained the expected sequence in all plants examined at 7 wpi ( S1 Table ) . To exclude the possibility that the truncations in the C-terminus of the RTD had any effects on the production of sgRNAs 3 from which ORF 7 would be translated , real-time quantitative PCR described by Hwang et al . , [25] was used to investigate the relative levels of PLRV sgRNAs 3 produced by WT virus and the deletion mutants . Primers for sgRNA 3 would detect genomic RNA as well as all three sgRNAs and comparable levels of RNA were detected in all samples ( S1E Fig ) . Primers for sgRNA2 would amplify genomic RNA as well as sgRNA1 and sgRNA2 , but not sgRNA3 . Similar to the sgRNA3 primers , these primers detected comparable levels of RNA in all samples ( S1E Fig ) , Taken together , these data support the conclusion that the difference in infectivity between Mut-Δ5670 and Mut-Δ5685 is not the result of altered transcription of sgRNA3 and ORF7 To determine if the 5-aa motif affected virus movement out of inoculated tissues , WT-PLRV , Mut-Δ5670 , Mut-Δ5685 , and Mut-Δ5670–5685 were individually infiltrated into HNS and N . benthamiana leaves . Tissue prints of HNS and N . benthamiana petioles from the infiltrated leaves were analyzed at 7 dpi and 5 dpi , respectively . Significantly more infection foci ( >10-fold ) were observed and counted in tissues of both hosts infected with WT-PLRV and Mut-Δ5685 ( i . e . those containing the 5-aa motif ) than in tissue infected with Mut-Δ5670 or Mut-Δ5670–5685 ( Fig 3 ) . As expected based on relative differences in virus titer between WT-PLRV and the 5-aa deletion mutants ( Fig 1B and 1C ) , the differences in numbers of infection foci between WT-PLRV and the 5-aa deletion mutants were smaller in N . benthamiana than in HNS ( Fig 3 ) . N . benthamiana plants infiltrated with Mut-Δ5670 and Mut-Δ5670–5685 expressed mild rather than the severe interveinal chlorosis symptoms observed in plants infiltrated with WT-PLRV and Mut-Δ5685 , and symptom expression was delayed 1 . 5–3 weeks ( S1A Fig ) . The nucleotides encoding the 5 aa ( LFEYQ ) would also be present in the ORF 7 encoded protein translated from sgRNA3 . To rule out that the defect in virus accumulation and symptom development of the 5-aa deletion mutants was due to changes in the ORF 7-encoded protein rather than the RTP , we constructed a mutant ( NoORF7 ) where the start codon of ORF 7 was eliminated by changing the “ATG” to “TTA” . The NoORF7 mutant virus accumulated to similar levels as WT-PLRV ( Fig 2D ) and typical interveinal chlorosis symptoms were observed by 3 wpi on HNS ( S1A Fig ) , indicating that elimination of the ORF 7 encoded protein had no effect on virus accumulation and symptom development . The NoORF7 mutant retained the expected sequence in all plants examined at 5 wpi ( S1 Table ) . Two of 30 HNS plants infected with Mut-Δ5670 , henceforth referred to as Rev1 and Rev2 , developed WT-PLRV symptoms at 10–12 wpi . Virus titer in Rev 1 and Rev 2 at 12 wpi was also comparable to WT-PLRV infected plants at 12 wpi ( S2A Fig ) . RT-PCR products spanning ORFs 5 , 6 , and 7 , and the 3’UTR ( nt 4215–5883 , Fig 1A ) were generated from Rev1 and Rev2 , cloned into pJET1 . 2 vector ( Fisher Scientific , USA ) , and sequenced . Ten clones were sequenced from each plant . All 10 clones from Rev1 and nine of 10 clones from Rev2 contained an 85-nucleotide insertion downstream of nt 5670 . This insertion was a duplication of the PLRV sequence encoded by nt 5588–5670 with an additional ‘tg’ at the 5’-terminus ( Fig 4A ) . The change in reading frame due to the addition of the “tg” resulted in the presence of a stop codon so that only 8 aa ( WFTADLNN ) would be added to the C-terminus of the truncated RTP ( Fig 4A ) . To obtain high-resolution virus population information in the Rev1 and Rev2 plants , targeted amplicon sequencing technique ( AmpSeq ) [46] was used to investigate the sequence diversity from tissues collected at 12 wpi . We found a dominant population ( 75% of the sequences ) that was exactly the same virus determined using Sanger sequencing ( Fig 4B ) . The remaining population contained the original Mut-Δ5670 ( 12% ) or insertions shorter than 85 nt , but all insertions were sequences within nt 5588–5670 ( S2B Fig ) . In addition to the insertion , there were several single nucleotide variations ( SNV ) along the RTD domain that differed between the wild-type and revertant virus populations ( S2C Fig ) . It is unknown if the SNVs in the revertant population were a result of the insertion , but those changes did not contribute to the alteration in movement or symptom expression phenotypes . In the revertant population there was a high sequence variation near nt 5670: At positions 5663 and 5671 there was a 40 . 0% and 83 . 6% C-T pyrimidine transition , respectively . At position 5672 there was a 94 . 31% T-G transversion . To determine if the eight-amino acid insertion found in both Rev1 and Rev 2 could restore the WT-PLRV phenotype , the 85 nt ( Rev-85nt ) and the 24 nt encoding the eight-aa sequence WFTADLNN ( Rev-8AA ) were re-constructed in the infectious PLRV clone and infiltrated into HNS plants . The number of virus infection foci in phloem tissue , quantified in petiole tissue prints developed from tissue 7 dpi , were similar for WT-PLRV , Rev-85nt , and Rev-8AA ( Fig 4C ) . All these viruses accumulated to similar levels , as measured by DAS-ELISA , in systemically infected HNS leaves in tissue collected at 5 wpi ( Fig 4D ) . Furthermore , all viruses induced interveinal chlorosis in HNS at ~4 wpi and were transmitted by aphids to HNS , P . floridana and potato at similar efficiencies indicating that the revertants were biologically similar to WT-PLRV ( S2 Table ) . The C-terminus of the PLRV RTD is predicted to be highly disordered [19] , however the disorder prediction algorithms IUPred and PONDR [47 , 48] identified ordered domains interspersed within the disordered region of the RTD ( S3 Fig ) . These ordered domains were predicted to be present in all the luteovirid sequences analyzed except in the enamoviruses , although the position of the ordered domains along RTD sequence varied among virus species ( S3 Fig ) . The 5-aa LFEYQ motif in the PLRV RTD we identified as being essential for efficient systemic spread and symptom development was associated with a conserved ordered domain . Alignments of the 5-aa motifs in 32 PLRV RTD sequences available in GenBank identified conservation of the three C-terminal amino acids ( EYQ ) , whereas there was some variation in the first two amino acids ( Fig 5A ) . Leucine was present in the first position of 27 isolates , the remaining five had a methionine ( M ) in the first position . The second position was occupied by one of three aromatic amino acids; Phenylalanine ( F ) was found in 25 isolates whereas six isolates contained a tyrosine ( Y ) , and one isolate contained a histidine ( H ) ( Fig 5A ) . Using the protein secondary prediction algorithms: I-TASSER ( https://zhanglab . ccmb . med . umich . edu/I-TASSER/ ) [49] , RaptorX Property Prediction ( http://raptorx . uchicago . edu ) [50] , and SPIDER2 ( http://sparks-lab . org/server/SPIDER2/index . php ) [51] , the last short ordered domain in the RTD C-terminus that was predicted between residues 450–500 was predicted to contain three α-helices ( S3 and S4 Figs ) . The first two residues ( LF ) of the 5-aa motif were located at the C-terminus of the terminal α-helix ( S4 Fig ) . None of the variations found in the first two amino acids would be predicted to disrupt the terminal α-helix ( S4 Fig ) . The sequence of the eight amino acid ( WFTADLNN ) insertion in the fully functional revertant virus was considerably different from the WT-PLRV sequence ( LFEYQ ) , only sharing the aromatic phenylalanine ( F ) residue in the second position , but it was predicted to form the terminal α-helix structure similar to WT-PLRV ( S4 Fig ) . Deletion of the LFEYQ motif ( Δ5aa , S4 Fig ) was predicted to completely abolish the terminal α-helix structure . Amino acids vary in their ability to form the various secondary structure elements; regions rich in alanine ( A ) , glutamate ( E ) , L , and M tend to form an α-helix , while proline ( P ) and glycine ( G ) residues tend to disrupt α-helices [52–54] . To determine whether the α-helix structure or/and the aromatic amino acid is/are important for RTP function , several mutants were constructed and classified into four categories based on their amino acid composition and structure as predicted by SPIDER2 ( S4 Fig ) . Category I mutants preserved the aromatic amino acid in position 2 and the predicted α-helix ( Mut-LYE , Mut-LHE , Mut-LFK , and Mut-MFE ) . Category II mutants were predicted to disrupt/shorten the terminal α-helix but would retain the aromatic amino acid ( Mut-PFG and Mut-GFP ) . Category III mutants were predicted to preserve the α-helix but would replace the aromatic amino acid ( Mut-LAE , Mut-LEE , and Mut-AAA ) . Category IV mutants were predicted to disrupt/shorten the α-helix and would also replace the aromatic amino acid ( Mut-PGE , Mut-LPG , and Mut-PGG ) . These 12 mutants were inoculated into HNS plants and virus systemic infection and symptom development were monitored ( Fig 5B ) . Only mutants from category I accumulated to levels comparable to WT-PLRV at 5 wpi . Mutants in categories II , III , and IV accumulated to low virus titer levels and no symptoms were expressed even at 8 wpi ( Fig 5B and 5C ) . All plants infected with category I mutants developed wild-type interveinal chlorosis symptoms , albeit the virus titer and the time for symptom appearance in plants varied among these mutants ( Fig 5B and 5C ) . Viral RNA was sequenced from representative plants infected by each mutant and all original mutations were maintained in the progeny virus ( S1 Table ) . Our previous studies have shown that PLRV CP/RTP exists as multiple isoforms in plants and that RTP monomers self-interact and interact with P17 [55 , 56] , a host-specific PLRV movement protein [8] . To test whether the 5-aa motif was required for these interactions , bimolecular fluorescence complementation assays ( BiFC ) were performed in agroinfiltrated fully mature ( source ) and expanding ( sink ) N . benthamiana leaves . Both wild-type RTP and Δ5aa-RTP ( expressed from Mut-Δ5670–5685 ) could self-interact , and the interaction signals were observed in the cytoplasm ( S5A and S5B Fig ) . Additionally , both Δ5aa-RTP and wild-type RTP interacted with P17 in the presence or absence of PLRV clone ( S5C–S5H Fig ) , indicating that the 5-aa motif had no observable effect on the protein-protein interactions between RTP monomers or between RTP and P17 . In the absence of replicating PLRV , the fluorescent RTP or Δ5aa-RTP and P17 complexes were observed in the cytoplasm and along the periplasm ( S5C–S5F Fig ) . In contrast , in the presence of PLRV , the fluorescent RTP-P17 complexes were observed mainly along the cell periplasm ( S5G Fig ) but the Δ5aa-RTP -P17 complexes were mainly in the cytoplasm ( S5H Fig and S1 Movie ) , indicating the 5-aa may affect RTP localization . Few fluorescent inclusions were observed in sink leaves when infiltrated with Yn-RTP/Yn-Δ5aa-RTP and Yc-P17 ( S5I–S5J Fig ) precluding any meaningful comparisons with localization in mature leaves . When expressed as a RTP-GFP fusion in N . benthamiana leaves PLRV RTP was localized in the cytoplasm and the nucleolus , but RTP lost its nucleolar localization in the presence of replicating PLRV [57] . To determine if the subcellular localization of RTP was affected by the 5aa deletion , the full-length wild-type RTP sequence and the RTP sequence from mutant Δ5670–5685 were fused with GFP ( both N and C terminal fusions were constructed ) and agroinfiltrated into fully mature ( source ) and expanding ( sink ) N . benthamiana leaves either alone or co-infiltrated with a full-length WT-PLRV infectious clone . These experiments were done in N . benthamiana leaves because infiltration into HNS resulted in a strong local HR reaction induced 1–2 dpi and interfered with our ability to study the effects of the PLRV mutants at the subcellular level . GFP-RTP was visualized by confocal microscopy at 3 dpi . When infiltrated independent of the infectious virus , both GFP-RTP and GFP-RTP-Δ5aa localized to the cytoplasm and nucleolus in both sink and source leaves ( Fig 6A , a-b and S6B Fig , a-b ) . When GFP-RTP was co-inoculated into mature leaves with the full-length infectious PLRV clone it localized to the cell periplasm and formed discontinuous , variable sized inclusion-like bodies ( ILBs ) along the cell wall/membrane ( Fig 6A , c-d , S6A Fig and S2 Movie ) . In contrast , when GFP-RTP-Δ5aa was co-inoculated with the infectious PLRV clone the ILBs were observed more frequently as cytoplasmic inclusions ( Fig 6A , e-f , and S3 Movie ) . To quantify the differences , the number and location of GFP-RTP and GFP-RTP-Δ5aa ILBs were determined in at least 20 infected cells ( Fig 6B ) . Z-Stacking of single planes through each infected cell was used to provide a composite image with a greater depth of field to better define the periplasm and a more accurate count of ILBs . Significantly more ILBs were observed in the cell cytoplasm of mature leaves co-infiltrated with GFP-RTP-Δ5aa + PLRV than those co-infiltrated with GFP-RTP + PLRV ( Fig 6B ) . Similar to results from the BiFC assays , the low number of ILBs were observed in sink leaves co-inoculated with the full-length infectious PLRV clone and either GFP-RTP or GFP-RTP-Δ5aa ( S6B Fig , c-d ) precluded any meaningful analysis . The PLRV P17 movement protein localizes to the plasmodesmata in mature tissues [58 , 59] . To determine if ectopically expressed P17 could direct RTP to the cell periphery , GFP-RTP was co-infiltrated with P17-mCherry into mature leaves . P17-mCherry localized to punctate spots , presumably plasmodesmata , along the cell wall when co-infiltrated with GFP-RTP ( Fig 7A–7C ) , but it did not redirect RTP from the cytoplasm to punctate spots along the cell wall ( Fig 7A–7C ) . When GFP-RTP , P17-mCherry and the PLRV infectious clone were all infiltrated together , RTP and P17 did co-localized to punctate spots along the cell wall ( Fig 7D–7F ) suggesting that additional virus factors or changes in the cellular environment induced by the virus were needed to direct RTP or a RTP-P17 complex from the cytoplasm to the cell periphery . To determine if the GFP-RTP inclusions observed along the cell wall were associated with plasmodesmata , GFP-RTP and the PLRV infectious clone were co-infiltrated with the plasmodesmata marker ( mCherry-PDLP1 ) [60] . Many of the GFP-RTP protein inclusions did co-localized with the mCherry-PDLP1 protein ( Fig 7G–7I ) . We observed 820 GFP-RTP ILBs in 12 different fields; 230 foci ( 28 . 04% ) co-localized with mCherry-PDLP1 and 422 ( 51 . 5% ) were adjacent to the mCherry-PDLP1 ( Fig 7G–7I ) . The high proportion of adjacent , but not perfectly overlapping mCherry-PDLP1 and GFP-RTP signals may be due to the dynamic nature of the GFP-RTP ILBs . Numerous inclusions were observed to traffic and accumulate together to form the ILBs ( S2 Movie ) . We also examined the localization of mCherry-PDLP1 and GFP-RTP co-infiltrated with the PLRV infectious clone in plasmolyzed N . benthamiana cells that would distinguish between punctate spots being associated with the cell wall or cell membrane [61] . Under these conditions , most of the GFP-RTP ILBs were co-localized with mCherry-PDLP1 ( Fig 7J–7L ) . In contrast , GFP-RTP-Δ5aa inclusions did not co-localize with the mCherry-PDLP1 protein in the presence of PLRV infectious clone ( S6C Fig ) . To determine whether the naturally modified RTP from the revertant virus also restored the wild-type RTP localization , the nucleotide sequence encoding WFTADLNN was cloned into the GFP expression construct in place of original amino acids ( GFP-RTP-Rev ) and infiltrated in N . benthamiana with infectious PLRV . GFP-RTP-Rev behaved similar to wild-type RTP-GFP with respect to subcellular localization in mature leaves ( Fig 8A , a and b ) . To determine if the aromatic amino acid and/or helix structure were responsible for RTP localization , mutant RTPs representing each of the four categories ( Mut-LYE , Mut-PFG , Mut-AAA , and Mut-PGE ) were cloned into the GFP conjugated vectors and ectopically expressed with the PLRV infectious clone in mature leaves . Mutations that preserved the aromatic amino acid and the helix structure , GFP-RTP-Rev ( Fig 8A and 8B ) and Mut-LYE ( Fig 8A , c ) formed ILBs at the periplasm similar to wild-type RTP ( Fig 8A , a ) . Mutations that replaced the aromatic amino acid and/or disrupted the helix structure ( Mut-PFG , Mut-AAA , and Mut-PGE ) resulted in ILBs being localized mainly in the cytoplasm ( Fig 8A , d-f ) . These results indicate that both the aromatic residue and the C-terminal α-helix located at the C-terminus of RTD domain are required for the RTP to localize properly in the cells to facilitate systemic movement .
Two forms of the polerovirus RTP are found in infected plants , a full-length soluble form and a C-terminal truncated form that is incorporated into the virion [19 , 26] . The RTD domain has a highly conserved N-terminal region and a variable C-terminal region . The N terminus is important for mediating RTP incorporation into the virion , aphid transmission , and interaction with aphid endosymbiont proteins [1 , 2] . Whereas the variable C-terminal region is dispensable for aphid transmission , it does play a role in tissue tropism , viral movement , virus accumulation , and symptom development in plant hosts , and it can facilitate efficient aphid transmission [2 , 17–20] . Furthermore , the C-terminus of RTD had no effect on RTD stability , but there are specific RNA motifs and structures located adjacent to the CP stop codon and 600 nt downstream that regulate RTP translation [62] . It was unknown whether specific RNA or protein domains or motifs can be associated with the different functions of the RTP . Here we identified that nucleotides 5671–5685 encode a protein motif near the C-terminus of the PLRV RTD that is required for efficient systemic infection and symptom development . Previous studies have reported that deletions or mutations in the RTD did not affect virus replication or accumulation in initially infected cells or at the cell level in systemically infected tissue whether the inoculum was 35S promoter-driven virus or an in-vitro transcript , but the number of infection foci in systemically infected tissues were reduced [11 , 18 , 19 , 21 , 62] . Our data also suggests that the 5-aa motif had no effect on virus multiplication and RTP accumulation ( Fig 2 ) . Deletion of nucleotides 5671–5685 reduced the number of infection foci in systemically infected leaves relative to WT-PLRV . This suggests that the 5-aa motif affected virus phloem loading in the inoculated leaves and/or unloading into the systemically infected tissues . While the requirement for stable virions for systemic infection of PLRV and related luteovirids is not host specific [3 , 63] , movement phenotypes and the role of movement-associated virus proteins , e . g . P17 and RTP , can be host specific . A requirement for the P17 of PLRV in virus movement was different in tobacco , P . floridana , and potato [8] . Deletion of RTD C-terminus in TuYV did not affect systemic infection of Montia perfoliata and A . thaliana; however , the mutant virus was unable to systemically infect N . benthamiana [20] . Similarly , mutations in the RTD C-terminus of PLRV affected the ability of the altered viruses to infect four different hosts [10] . Therefore it was not surprising that we found the phloem loading phenotypes of the PLRV 5-aa mutants differed among the four hosts used in this study . The most significant effects were in the natural hosts ( potato and HNS ) and less so in N . benthamiana and P . floridana . The significant restriction on movement in the phloem of HNS resulted in a strong selection pressure and the generation of revertants that recovered RTP function allowing efficient movement in the phloem and systemic infection . While systemic infection was reduced in N . benthamiana and P . floridana , it was not as significant as in HNS and therefore the selection pressure to generate revertants was not likely as strong . We did not attempt to detect virus mutations in potato due to a low agroinfiltration efficiency and longer times to develop a systemic infection in the few plants that did ultimately become infected . Naturally generated compensatory mutations have been reported from several studies when engineered deletions or mutations in the RTP coding region have negatively impacted local or systemic movement . Brault et al . [16] , found pseudo-revertants of the Beet western yellows virus ( BWYV ) RTP restored the reading frame and the ability of the virus to move efficiently in phloem and to accumulate to wild-type levels in plants . Similarly , a mutation resulting in a frameshift and truncation of the PLRV RTD that altered the tissue tropism led to the selection of intragenic mutations or pseudorevertant mutations , either nucleotide insertions or larger deletions . All of these additional mutations resulted in an additional frameshift that restored the translation of the P5 C terminus and the WT-PLRV infection phenotype [26] . A point mutation in the BWYV RTD that impeded virus accumulation was found to either revert back to the wild-type sequence or pseudo-revert to an amino acid that was functional . Aphid transmission that was affected by other single site mutations was restored by compensatory mutations at other sites [11] . The selection of revertants or pseudo-revertants that restore virus movement and systemic infection has also been reported in several unrelated plant viruses further suggesting that movement function is under strong selection pressure . Thompson et al . [64] , found compensatory capsid protein mutations in cucumber mosaic virus conferring systemic infectivity in Cucurbita pepo . Second-site reversion of a dysfunctional mutation in a conserved region of the tobacco mosaic virus movement protein was found to restore systemic infection [65] . Maize chlorotic mottle virus p31 protein , expressed as a readthrough extension of p7a , is required for efficient systemic infection . Pseudorevertants were detected to restore MCMV systemic infection when a stop codon was introduced downstream of p7a stop codon that disrupted p31 expression [66] . The variable C-termini of luteovirid RTDs are highly disordered [2 , 19 , 23] , and therefore their structure is not well studied . We used several algorithms that independently predicted several helix structures in the C-terminus of the RTP to be conserved across all available PLRV sequences . Similar analysis of other luteovirids , except the enamoviruses which do not have the equivalent of the C-terminal RTD domain , predicted similar conserved structural motifs , albeit the positions vary among viruses ( S3 and S7 Figs ) . The pseudo-revertants of Mut-Δ5670 that restored RTP translation and movement function were generated by duplication of sequences from a specific genome region , all of which would be predicted to recover the α-helix structure . Taken together these data support our hypothesis that the ordered α-helix structure plays an important role in RTP intrinsic functions and/or RTP-host protein interactions , and that mutations to maintain the structures will be selected . While α-helices can form helix-turn-helix motifs , leucine zipper motifs or zinc finger motifs that have particular significance in DNA binding [67–69] , none of these structures were predicted in the RTD using COILS and Prabi algorithms ( S8 Fig ) [70] . The α-helix is also the most common structural element in transmembrane proteins [71] . The Δ5aa-RTP could not move efficiently to the cell periplasm and was sequestrated in the cytoplasm suggesting that RTP trafficking may depend on the plant endomembrane system , and the conserved C-terminal α-helix may be required for anchoring the RTP to the endomembrane . Alternatively , the α-helix may be a transmembrane domain that associates with specific lipids and this lipid-protein interaction determines the protein localization [72] . Protein structure-mediated intracellular virus trafficking is common . The NSm protein of groundnut bud necrosis virus interacts with endoplasmic reticulum membranes via a coil-coil domain at the C-terminus to remodel the endoplasmic reticulum network to form vesicles that in turn translocate NSm to plasmodesmata [73] . Deletion of 66 amino acids from the C-terminus of the tobacco mosaic virus movement protein resulted in an inability of the protein to associate with microtubules and move to plasmodesmata near the leading edge of infection [74] . Protein function is often dependent on protein structure that is determined by the composition and chemistry of its amino acids [75] . Aromatic-aromatic interactions are ubiquitous in nature and play important roles in maintaining structures of proteins [76] . For example , aromatic residues in the C-terminal helix of human apoC-I mediate phospholipid interactions and particle morphology [77] . The conserved aromatic residue in the C-terminal α-helix appears to be a particular structural feature that enables RTP to function in virus movement . The predicted structural analysis coupled with changes in biological functions we observed support a conclusion that the C-terminal domain of luteovirid RTDs would be better described as having important ordered structures at both ends with an intrinsically disordered central region . While the disordered regions will preclude structural analyses of the RTD using available technologies , we have been able to show that conserved structures in this multifunctional protein are important and this lays the groundwork for future studies . To the best of our knowledge , this is the first report that defines specific structural motifs in the C-terminus of the RTD and shows the importance of the aromatic residue in the C-terminal α-helix for PLRV phloem loading . Poleroviruses traffic systemically as virions and the P17 movement protein does associate with the PLRV virion [3 , 55] . Whether there is an association between P17 and RTP was unknown . Here we found RTP formed numerous ILBs along the cell periphery in mature source leaves , but not in developing sink leaves when co-inoculated with the PLRV infectious clone ( S6A and S6B Fig ) . Moreover , while the RTP also interacted with P17 in the developing sink leaves in our BiFC test , the number of ILBs was much lower and the signal was less consistent than that in source leaves ( S5I–S5L Fig ) . This is likely explained by previous research showing that P17 localizes to branched plasmodesmata in mature source leaves and roots , but not to the simple plasmodesmata found in developing sink leaves where P17 is degraded by the 26S proteasome [58 , 59 , 78] . Collectively , it appears that while RTP localization in mature N . benthamiana leaves requires an interaction with P17 , other virus encoded factors are contributing to the subcellular localization of the RTP since ectopically expressing P17 alone was not enough to target RTP to plasmodesmata . The recently discovered polerovirus P3a protein is required for efficient virus infection [9] , but its role at the cellular and subcellular level with virus infection is unknown . Alternatively , host components present in virus infected cells may play a role in facilitating P17-RTP interactions and localization . The plasmodesmal permeability is considerably down-regulated during the sink-source transition in leaves , which is accompanied by a progressive conversion from simple to branched forms of plasmodesmata . The lack of RTP localization at plasmodesmata in sink leaves suggests a differential role of the RT protein in source and sink leaves . The manner in which poleroviruses induce interveinal chlorosis symptoms in their hosts is not well understood . In general , this type of symptom has been attributed to increased photoassimilate retention and starch accumulation in the source leaves which blocks starch transport to phloem [79] . The accumulation of virions or overexpression of virus-encoded proteins can cause a reduction in plasmodesmata permeability to photoassimilates by the excessive accumulation of callose at the plasmodesmata [18 , 80 , 81] . Rinne et al . [81] , showed that constitutive expression of the tospovirus movement protein NSm in tobacco induced virus infection symptoms . This was correlated with the obstruction of NSm-targeted mesophyll plasmodesmata in source tissues by depositions of 1 , 3-beta-D-glucan or callose . Bruyere et al . [18] , observed an absence of symptoms following agroinfiltration of N . clevelandii plants with BWYV containing a deletion of the C-terminal RTD region . They also reported two deletions in the N- terminus of BWYV RTD abolished symptoms . These two domains appear to correspond to the two domains we recently found to regulate RTP translation [62] . The loss of symptom expression associated with the deletions in the N-terminus of the BYWV RTD was likely an indirect effect due to a loss of RTP translation not a direct effect on symptom development . Furthermore , large deletions in the N-terminus of the PLRV RTD that do not interfere with RTP translation do not affect symptom expression ( S3 Table ) . Our observation that PLRV RTP formed ILBs located at the cell periphery and associated with plasmodesmata in infected mature source leaves suggests that the RTP , perhaps in concert with other virus proteins , when bound at plasmodesmata can reduce the size of pore plasmodesmata units and reduce permeability to photoassimilate . ILBs comprised of RTP and other virus proteins localized at the cell periplasm and plasmodesmata of mature leaves may interact with host factors through the C-terminal domain and regulate the plasmodesmata permeability and perturbation of the symplastic connectivity for photoassimilate by inducing callose accumulation . This hypothesis is supported by a recent finding that a protein kinase binding to the C-terminal domain of the RTP of turnip yellows virus regulates virus accumulation . This protein kinase was found to localize to the plasmodesmata [20] . The composition of the RTP-associated punctate bodies associated with plasmodesmata and their role in regulating the size of pore-plasmodesmata needs to be investigated in more detail . Both structural and non-structural forms of the RTP of Cucurbit aphid-borne yellows virus are essential for efficient systemic infection [19] . Here we show that in the absence of the full-length RTP , PLRV is deficient in phloem loading and systemic infection . The requirement for the C-terminal domain of the PLRV RTP for efficient long distance movement is supported further by studies on the related member of the Luteoviridae , Pea enation mosaic virus ( PEMV ) that lacks the homologous C-terminal part of the RTP and it is unable to move systemically in the absence of the helper umbravirus PEMV-2 [82] . Our data strongly support the hypothesis that efficient long-distance movement of poleroviruses requires the C-terminus to help with efficient phloem loading of virions . Indeed , disrupting virus phloem loading and systemic infection are common resistant strategies that plants adopt to counter virus infection [30] . Several naturally occurring host resistant genes , which restrict/promote virus phloem loading and long-distance movement have been identified [83–86] . Identifying the virus and host components involved in the interactions with the C-terminus of RTP and its role in cellular and systemic movement , along with an understanding of how the virus can compensate for perturbations of the RTP sequence should assist in developing sustainable host resistance to systemic infection .
Recombinant cDNA of PLRV mutants used in this study were generated using specifically designed primers according to the previously described yeast recombination method [87] . All the primers are available upon request . Briefly , by using a set of restriction enzymes , BsrG1 and Xho1 or BsrG1 and Bsa1 , the plasmid pBPY , a yeast-bacteria shuttle vector containing a full-length cDNA of PLRV , was linearized and dephosphorylated with alkaline phosphatase ( New England Biolabs , Ipswich , USA ) to inhibit recircularization . The short insertion fragments containing the mutated region were PCR-amplified from pBPY with primers homologous to their junctions . Yeast transformation was carried out based on the LiAc/SS carrier DNA/PEG method [88] . Total DNA including the recombinant plasmid DNAs were rescued from pooled yeast colonies grown on the SD/-Trp media agar plate ( Clontech , Mountain View , CA ) , and then used to transform Escherichia coli strain DH5α . Plasmid DNAs recovered from DH5α were sequenced to verify sequence integrity . Agrobacterium tumefaciens strain LBA4404 was transformed with plasmids extracted from DH5α and used for agroinfiltration of N . benthamiana , S . sarrachoidies ( HNS ) , P . floridana and S . tuberosum as previously described [28] . To construct BiFC vectors , WT-RTP , Δ5aa-RTP and P17 were fused to the N-terminal or C-terminal fragment of YFP in the Yn and Yc vectors , respectively as described [89] , and then were used for agroinfiltration into N . benthamiana . Primers used in this study are listed in S4 Table . All N . benthamiana plants used in this study were grown in greenhouse maintained at 25 ± 2°C with a 16:8 h photoperiod , and 4–6 leaves were used for infiltration . Agroinfiltrated N . benthamiana leaf tissues harvested 3 days post infiltration ( dpi ) were ground to a powder in liquid nitrogen using a pestle and mortar; 100 mg was resuspended in 0 . 5ml of cold extraction buffer containing 50 mM Tris-HCl pH 6 . 8; 2 . 5% SDS; 10% glycerol and 100 mM DTT ( added fresh ) and incubated on ice for ~20 minutes with occasional vortexing . Samples were centrifuged , 16 , 000 g for 5 min , and 50 ul of the supernatant was diluted 1:1 in 2×Laemmli sample buffer ( Bio-Rad , Hercules , CA ) supplemented with BME ( 5% v/v ) . For Western blot , samples were incubated at 100°C for 8–10 min , separated on a 10% Mini-PROTEAN TGX precast gel ( Bio-Rad ) , and transferred to nitrocellulose . SDS-PAGE , transfer and Western blotting were performed as described [56] . PLRV CP and RTP were detected using by using a monoclonal antibody , SCR3 , that recognizes the N-terminus of the CP [90] . Blotted signal was visualized using chemiluminescence according to the manufacturer’s manual ( GE Healthcare , USA ) and signal band quantification was measured with ImageJ software ( https://imagej . nih . gov/ij/ ) . All experiments were repeated at least three times . Amino acids encoded from luteovirids RTP genes ( ORF3/5 ) were used to predict the disorder domains by using the disorder prediction algorithms IUPred [48]and PONDR [47] . The protein secondary structure was predicted using I-TASSER ( https://zhanglab . ccmb . med . umich . edu/I-TASSER/ ) [49] , RaptorX Property Prediction ( http://raptorx . uchicago . edu ) [50] , and SPIDER2 ( http://sparks-lab . org/server/SPIDER2/index . php ) [51] . Alignment of luteovirid sequences was performed using an online version of CLUSTALW 2 . 1 software with default “slow/accurate” parameters ( http://www . genome . jp/tools/clustalw/ ) . Graphical representation of the sequence conservation of 32 PLRV RTD sequences available online was generated by using the online software WebLogo [91] . Virus accumulation in agroinfiltrated leaves and systemically infected plants ( three leaves and fifteen plants ) was detected by using double-antibody sandwich enzyme-linked immunosorbent assay ( DAS-ELISA ) as described previously [4] . Total RNA was extracted from N . benthamiana leaves 3 dpi as described previously [92] . PLRV gRNA and subgenomic RNA1 were detected by Northern blot using DIG Northern Starter kit per the manufacturer’s instructions ( Sigma , USA ) with minor modifications . RNA gel and blot analysis were performed using a protocol adapted from Xu et al [92] . Thirty-five micrograms of nucleic acid from each sample were separated on a 1% agarose–6% formaldehyde gel . A 300 bp DNA probe covering the PLRV genome from position 3820 to 4119 was obtained by PCR and used as template for probe labelling with Digoxigenin-dUTP . The complementary DNA strand of denatured DNA was synthesized by Klenow polymerase using the 3′-OH termini of the random oligonucleotides as primers . Progeny viruses in systemically infected plants were analyzed by sequencing reverse transcription-PCR ( RT-PCR ) products . Tissue immunoblot analysis was performed on petioles of leaves infiltrated with the various mutants at 7 dpi on HNS , and 5 dpi on N . benthamiana as described previously [4] . Briefly , the nitrocellulose membranes were treated with 0 . 2 M CaCl2 prior to blotting . Tissue prints were allowed to air dry and stored at 4°C until the end of the experiment , at which time the samples were analyzed under the same conditions . PLRV coat protein was detected in tissue blots using anti-PLRV immunoglobin and goat anti-rabbit immunoglobulin conjugated to alkaline phosphatase ( 1:5000 ) as secondary antibody ( Promega , USA ) . The presence of alkaline phosphatase was detected using the 1-Step NBT/BCIP system ( Pierce , Rockford , IL ) . To visualize the subcellular localization of RTP alone or in the presence of infectious PLRV , leaves of 6-week-old N . bethamiana plants were infiltrated with A . tumefaciens ( strain GV3101 ) harboring either the full length GFP-RTP sequence or one of the GFP-RTP ( Δ5aa ) plasmids with or without co-infiltration of A . tumefaciens harboring a plasmid containing the full length PLRV clone as described previously [28] . Agrobacterium suspensions at 0 . 8 optical density ( OD600 ) was used for infiltration . To visualize intracellular structures , the plasmodesmata marker plasmid , mCherry-PDLP1 ( Sergey Ivanov et al . , 2014 ) , was transformed into A . tumefaciens GV3101 and co-infiltrated with two additional A . tumefaciens cultures transformed with GFP-RTP and the PLRV infectious clone . Leaf tissue was harvested at 48h post agroinfiltration and examined for GFP fluorescence using a Leica TCS-SP5 confocal microscope ( Leica Microsystems , http://www . leica-microsystems . com/ ) with a 20× objective . Agroinfiltrated N . benthamiana leaves were collected at 3 dpi for GFP observation in tissues that were co-infiltrated with the fusion proteins and PLRV infectious clone . Z-stack acquisition intervals were selected to satisfy Nyquist sampling criteria . We calculated the number of spots from 30 single sections from each cell , and the membrane targeting was strictly controlled on single sections . The total number of spots was the sum of the distinct spots from each of these 30 single sections that make up the stacks . Conditions set to excite GFP and monitor the emission were as described by Ivanov et al [60] . Confocal images were processed using Leica LAS-AF software ( Leica Microsystems ) . The GFP fluorophore was excited with a 488 nm laser and emission was detected at 505–545 nm . The RFP fluorophore was excited with a 532 nm laser and emission was detected at 588 nm . The segment of the PLRV genome spanning the entire RTD domain ( nt 4179–5883 ) was cloned by RT-PCR from WT-PLRV infected HNS and the revertant plants at 12 wpi , respectively . Library construction and next-generation sequencing were performed at the Cornell University Biotechnology Resource Center as described by Yang et al [46] . The raw data were aligned by Burrows-Wheeler Aligner ( BWA ) software [93] after quality check and PCR duplication filter . The reads were visualized by Tablet [94] . Single nucleotide variations ( SNV ) along the RTD domain were generated by Lofreq ( version 2 ) using default settings [95] . Virus-derived reads from WT-PLRV and the revertant virus library were mapped to the reference genome ( wild-type RTD sequence after removing PCR duplicates ) to mitigate the effects of PCR amplification bias introduced during library construction . Real-time quantitative PCR was performed based on Hwang et al . , [25] with modifications . Briefly , total RNA was extracted individually from N . benthamiana leaves infiltrated with WT-PLRV , Mut-Δ5670 or Mut-Δ5685 at 3 dpi by using TRIzol reagent ( Invitrogen , USA ) as described above . Contaminating DNA was removed by using DNase I treatment according to the manufacturer’ instructions ( Ambion , USA ) . cDNA was synthesized using iScript cDNA Synthesis Kit ( Bio-Rad , USA ) according to the manufacturer’s instruction . Primers were designed using Primer Premier 6 software ( Premier Biosoft , Palo Alto , USA ) , and PLRV sgRNA2 primer set ( P2 ) was F: 5’-ATGCTACAGTCGATGGTACAGA-3’ and R: 5’-CTATAATAGTAGCAGCAGCGACAT-3’; PLRV sgRNA3 primer set ( P3 ) was F: 5’-TGGAGAAGAGAGAGGAAAATGTC-3’ and R: 5’-TCTTAAGGGTCCCTCCCTCGA-3’ . RT-qPCR was performed using CFX96 Touch Real-Time PCR Detection System with the SsoAdvanced Universal Probes Supermix ( Bio-Rad ) . The N . benthamiana actin gene was used as internal standard , and 1 ng plasmid containing the full-length of PLRV infectious clone was used as control for normalization . Each experiment was performed with three biological replicates and three technical replicates . HNS plants systemically infected with WT-PLRV or the 5-aa deletion mutants were used as virus source tissue in aphid transmission assays . The tests were performed as described previously by Peter et al . [10] with the exception that eight aphids were placed onto each recipient plant . A 24 h acquisition access period was followed by a 5-day inoculation access period . The fumigated plants were placed in a greenhouse and observed for symptom expression and assayed for systemic infection by double antibody sandwich ( DAS ) -ELISA at 3–5 wpi . | Protein function is often dependent on its structure that is determined by the composition and chemistry of its amino acids . The C-terminal ~200 amino acids of the PLRV RTP are characteristically disordered yet this protein domain is involved in virus movement , tissue tropism and symptom development . Analysis of virus mutants , virus populations , and their cell biology allowed us to identify a short-ordered stretch of amino acids containing a conserved aromatic amino acid in the disordered RTP C-terminus . This structural feature determines efficient long-distance virus movement and symptom expression in a host-dependent manner . Virus mutants deficient in this feature generate compensatory mutations by duplication of their own sequence that restore the aromatic amino acid and the associated α-helix structural motif and in turn restore the wild-type symptom and movement phenotypes . Understanding the adaptive abilities of the virus should help to design stable virus constructs and sustainable host resistance strategies . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] | [
"plant",
"anatomy",
"chemical",
"compounds",
"plant",
"cell",
"biology",
"vascular",
"bundles",
"microbiology",
"phloem",
"plant",
"physiology",
"cloning",
"viral",
"structure",
"organic",
"compounds",
"plant",
"science",
"amino",
"acids",
"sequence",
"motif",
"analysis",
"molecular",
"biology",
"techniques",
"cellular",
"structures",
"and",
"organelles",
"research",
"and",
"analysis",
"methods",
"sequence",
"analysis",
"aromatic",
"amino",
"acids",
"plasmodesmata",
"bioinformatics",
"proteins",
"chemistry",
"leaves",
"molecular",
"biology",
"virions",
"cytoplasm",
"biochemistry",
"plant",
"cells",
"cell",
"biology",
"organic",
"chemistry",
"virology",
"database",
"and",
"informatics",
"methods",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"physical",
"sciences"
] | 2018 | An aromatic amino acid and associated helix in the C-terminus of the potato leafroll virus minor capsid protein regulate systemic infection and symptom expression |
We have used Digital Gene Expression analysis to identify , without bilaterian bias , regulators of cnidarian embryonic patterning . Transcriptome comparison between un-manipulated Clytia early gastrula embryos and ones in which the key polarity regulator Wnt3 was inhibited using morpholino antisense oligonucleotides ( Wnt3-MO ) identified a set of significantly over and under-expressed transcripts . These code for candidate Wnt signaling modulators , orthologs of other transcription factors , secreted and transmembrane proteins known as developmental regulators in bilaterian models or previously uncharacterized , and also many cnidarian-restricted proteins . Comparisons between embryos injected with morpholinos targeting Wnt3 and its receptor Fz1 defined four transcript classes showing remarkable correlation with spatiotemporal expression profiles . Class 1 and 3 transcripts tended to show sustained expression at “oral” and “aboral” poles respectively of the developing planula larva , class 2 transcripts in cells ingressing into the endodermal region during gastrulation , while class 4 gene expression was repressed at the early gastrula stage . The preferential effect of Fz1-MO on expression of class 2 and 4 transcripts can be attributed to Planar Cell Polarity ( PCP ) disruption , since it was closely matched by morpholino knockdown of the specific PCP protein Strabismus . We conclude that endoderm and post gastrula-specific gene expression is particularly sensitive to PCP disruption while Wnt-/β-catenin signaling dominates gene regulation along the oral-aboral axis . Phenotype analysis using morpholinos targeting a subset of transcripts indicated developmental roles consistent with expression profiles for both conserved and cnidarian-restricted genes . Overall our unbiased screen allowed systematic identification of regionally expressed genes and provided functional support for a shared eumetazoan developmental regulatory gene set with both predicted and previously unexplored members , but also demonstrated that fundamental developmental processes including axial patterning and endoderm formation in cnidarians can involve newly evolved ( or highly diverged ) genes .
A major challenge in biology is to understand how the current extraordinary diversity of animal forms has been generated during evolution . Specific goals are to determine which genes were employed to regulate developmental processes in the earliest multi-cellular animals , and how this set of regulators was expanded during the evolution of different animal branches by diversification of existing gene families or by the acquisition of new genes . To address these questions requires identification and functional analysis of developmental regulatory genes in species from right across the animal kingdom , covering not only the “bilaterian” ( protostome plus deuterostome ) branch including the classic experimental models such as mouse , zebrafish , Drosophila and Caenorhabditis , but also non-bilaterian phyla such as cnidarians , ctenophores and sponges , which have evolved many distinct forms and body plans . Following the recent explosion of genome and transcriptome sequencing it has been widely noted that the majority of families of transcription factors and signaling pathway components uncovered as developmental regulators in bilaterian model species are represented in genomes of cnidarians , as well as ctenophores and to a lesser extent sponges [1]–[9] . This has fuelled the idea that a shared set ( or “common toolkit” ) of genes inherited from a common metazoan ancestor is used to regulate development in widely divergent species through differential deployment [1] , [3] , [10]–[17] . The common toolkit idea relies heavily on the assumption that conserved genes have retained largely equivalent developmental functions during the evolution of each animal lineage , for which evidence remains quite patchy . Comparing the expression territories , and in some cases functions , of gene orthologs in families of transcription factors such as Hox , Sox , Fox and T-box genes , and components in signaling pathways such as Wnt , TGFβ , FGF , Hedgehog , Notch etc , has provided some support for this assumption , but also for lineage specific modifications in gene repertoires for example through gene duplications and losses within the transcription factor families [10]–[21] . Another possibility is that novel regulatory genes have emerged within specific evolutionary lineages to contribute to generating animal diversity [14] , [18]–[23] . The significant proportions of cnidarian-specific gene sequences in the fully sequenced genomes of Hydra ( around 15% ) and Nematostella ( around 13% ) is compatible with such a scenario in Cnidaria [14] , [18] , [22]–[26] . Detailed studies involving transcriptome comparisons in Hydra have shown that many cnidarian-specific genes are associated with specialized cell types , notably nematocytes ( stinging cells ) but also nerve and gland cells [22] , [24]–[30] , while others have been specifically implicated in intercellular signaling and regulating morphological processes [22] , [27]–[31] . Furthermore , in a subtractive hybridization search for cnidarian-specific genes involved in embryogenesis , 30 of 88 distinct partial cDNA clones recovered did not match known bilaterian sequences , including one corresponding to a Hydra specific gene ( HyEMB-1 ) expressed in the ovary and early embryo [31] . To gain a fresh perspective on the gene repertoires that regulate metazoan development , we employed a systematic unbiased comparative transcriptomics approach to identify potential regulators of embryonic patterning at gastrula stage in the cnidarian experimental model Clytia hemisphaerica [32] . Clytia is a typical hydrozoan species that includes a jellyfish form as well as a polyp form in its life cycle , unlike anthozoan cnidarians such as the popular sea anemone model Nematostella vectensis . After gastrulation , a torpedo-shaped “planula” larva is formed , whose organization shows the characteristic cnidarian body plan: a single “oral-aboral” axis and two germ layers . The outer ectoderm of the Clytia planula features ciliated epitheliomuscular cells for motility , and an internal endodermal ( or “entodermal” ) region including a population of interstitial stem cells ( i-cells ) specific to hydrozoans , which generate a variety of cell types for each germ layer [33]–[36] . Gastrulation proceeds by unipolar cell ingression to fill the blastocoel prior to endoderm cell epithelialization [37] . The gastrulation site derives from the egg animal pole and corresponds to the pointed oral pole of the larva , giving rise after metamorphosis to the mouth region of the polyp form [38] . Establishment of the oral pole in Clytia critically depends on Wnt/Fz signaling activity through the Wnt/β-catenin pathway . Maternally-provided transcripts for the ligand Wnt3 and the receptors Fz1 ( activatory ) and Fz3 ( inhibitory ) are pre-localized along the egg animal-vegetal axis to drive activation of this pathway on the future gastrulation site/oral side during cleavage and blastula stages [39] , [40] . This activation establishes distinct regional identities characterized by specific sets of transcribed genes at the oral and aboral poles of the developing embryo , including those required for cell ingression at gastrulation . Fz-PCP signaling , dependent on the conserved transmembrane protein Strabismus ( Stbm ) , is activated in parallel along the same axis to coordinate cell polarity in the ectoderm and to guide embryo elongation [41] . Since multi-member Wnt families with early polarized embryonic expression have also been uncovered in other cnidarians [42] , [43] , ctenophores and sponges [44]–[47] as well as in a range of bilaterian models [48] , [49] , it seems highly probable that Wnt/Fz signaling regulated embryonic patterning in ancestral metazoans , specifying the primary body axes and/or presumptive germ layer regions . To identify genes potentially involved in Clytia embryogenesis without favoring gene families identified as developmental regulators from bilaterians , we compared transcriptomes at the onset of gastrulation between normal embryos and ones strongly “aboralized” by Wnt3 morpholino ( Wnt3-MO ) injection prior to fertilization [40] . In many animals gastrulation coincides with , or closely follows , a significant stepping up of transcription from the zygotic genome , taking over from an initial phase of development predominantly dependent on maternally supplied mRNAs and proteins . By comparing transcriptomes from undisturbed and Wnt3-MO early gastrulae by Digital Gene Expression ( DGE ) we compiled lists of significantly over- and under-expressed genes . These included orthologs of known conserved developmental regulators but also members of unexplored metazoan conserved gene families , and in addition many sequences restricted to cnidarians . Expression profiling for an unbiased subset of these transcripts systematically revealed spatially or temporally restricted expression profiles of four types . Further transcriptome and in situ hybridization comparisons with Fz1-MO and Stbm-MO embryos revealed expression-pattern-related differences in the responses of genes to disruption of Wnt/β-catenin versus PCP . Finally , roles in developmental processes for the identified genes , both conserved and cnidarian–restricted , were supported both by their characteristic expression patterns and by correlated phenotypes obtained following morpholino injection for a subset of 8 genes . Overall our unbiased screen allowed systematic identification of developmental genes regulated by the Wnt/ß-catenin pathway and by Fz-PCP . It provided functional support for a shared eumetazoan developmental regulatory gene set with both predicted and previously unexplored members , while also showing that axial patterning and endoderm formation in cnidarians can involve taxon restricted genes .
To identify genes regulated transcriptionally in relation to Wnt dependent embryo patterning we compared transcriptomes from unmanipulated early gastrula stage embryos and from embryos injected prior to fertilization with a morpholino antisense oligonucleotide targeting Wnt3 [40] . Digital Gene Expression analysis ( DGE ) was performed using an Illumina HiSeq sequencing platform . The number of mapped reads onto a reference transcriptome data set was taken as a measure of transcript level , and the statistical significance of differences in these levels between samples assessed using the DEGseq package ( Figure 1; see Materials and Methods for technical details ) . Plotting for each transcript the expression ratio between two samples against the global average expression ( Figure 1A , C ) allowed visualization of sets of transcripts that showed significant differential expression , defined as ones that cannot be accounted for by sampling variation according to Random Sampling Model . We used the MATR method [50] , justified by the Normal distribution of the data ( Figure 1B ) , to adjust the cutoff to take into account experimental noise , based on comparison of replicate samples ( blue line in Figure 1A: compare with the red line delimiting the theoretical random distribution ) . For subsequent analyses we routinely used a corresponding “z-score” value as an index of significant differences between samples ( see Methods ) . Comparisons between Wnt3-MO and uninjected embryo samples ( Figure 1C ) identified 375 assembled transcript sequences as differentially expressed according to the z-score +/−3 . 3 cutoff , which corresponds to a probability threshold ( p-value ) of 0 . 01 ( colored dots in Figure 1C ) . Detailed analyses were performed for a more restricted set of 179 sequences with z-scores of less than -5 or greater than +5 ( see insert in Figure 1C; list of transcripts and their characteristics in File S1 ) . We could eliminate transcripts whose expression levels were affected non-specifically by the morpholino injection procedure by comparing the Wnt3-MO embryo differentially regulated transcripts with those identified in embryo populations generated using morpholinos targeting two other genes , Fz1 and Fz3 , which respectively activate and repress Wnt/β-catenin signaling leading to aboralized and oralized phenotypes , respectively [39] . Genes non-specifically affected by the morpholino injection procedure are expected to respond in the same way in all three experimental groups , whereas genes regulated specifically downstream of Wnt3 are expected to respond distinctly following Fz1-MO compared to Fz3-MO injection . Comparison between these groups allowed us to identify 4 sequences with high z-scores ( >5 ) in Fz1-MO and in Fz3-MO ( opposite phenotypes ) as well as Wnt3-MO samples ( purple dots in Figure 1D , E; DGE class 5 in File S1 ) . Two of these code for Ubiquitin ligases , implicated in protein degradation , and one for a secreted cyclase , suggesting a possible association with lysis of damaged cells in injected embryos . In addition the Fz3 transcript was itself detected at high levels in Fz3-MO embryos , probably due to the stabilizing effect of the morpholino . An additional set of 10 transcripts were eliminated as coming from likely bacterial contaminants , because they clearly stood apart as strongly under-represented ( z-scores <−5 ) in both Fz3-MO and Wnt3-MO samples ( and also for Fz1-MO in 9 cases ) compared with uninjected controls ( blue dots in Fig 1D , E . ) . The sequences of these transcripts had no similarity with any known eukaryotic genes but rather included genes from bacteria . Contamination from bacteria may be higher in uninjected embryos due to reduced manipulation of the egg and thus more frequent retention of the jelly coat and associated contaminants . After elimination of the 13 non-specifically affected sequences , our final validated transcriptome comprised 166 differentially expressed transcripts . 153 of these 166 had clear predicted full or partial ORFs , comprising 40 over-expressed in Wnt3-MO embryos and 114 under-expressed . Detailed analysis of these sequences ( File S1 ) revealed conserved and novel genes . We undertook detailed characterization of spatial expression and sequence analysis ( Table 1 ) non-selectively for the top 20 under-expressed transcripts ( Figure 2 ) and top 18 over-expressed transcripts ( Figure 3 ) in Wnt3-MO early gastrulae . Expression territories for all 38 transcripts were determined by in situ hybridization at three stages: early gastrula , 24 hpf planula ( just completed gastrulation , endoderm still undifferentiated ) , and 48 hpf old planula ( cell differentiation ongoing in both endodermal and ectodermal regions ) . We found that almost all the in situ hybridization profiles could be assigned to one of four types , which we termed Oral ( O ) , Aboral ( A ) , Ingressing/Endodermal ( IE ) and Delayed expression ( D ) types , as described in more detail below and summarized in Figure 4 . Briefly , O and A type profiles are characterized by polarized expression with respect to the developing oral-aboral axis at all stages , suggesting ongoing patterning roles during embryonic and larval development . The IE type profile corresponds to cells destined to contribute to the complex endodermal region including the i-cell stem cells and their derivatives . The D type profile transcripts were barely detectable in early gastrulae but showed at larval stages expression in diverse patterns in the ectoderm and/or later in the endoderm . Overall , our approach to identify new candidates for roles in cnidarian embryonic development was completely validated by these analyses . Without any selection based on sequence identity , all the transcripts we tested showed expression restricted in space and/or time during gastrulation and planula development . Names were assigned to the analyzed transcripts on the basis of orthology and/or membership of known gene families ( all phylogenetic analyses in File S2 ) . Multiple members of known gene families were distinguished by suffixes designating the 4 main expression profile types: O , A , IE or D . Cnidarian-specific transcripts lacking any recognizable orthologs from non-cnidarian species in NCBI databases , and those with non-cnidarian orthologs that had not previously been characterized , were assigned novel names using the same suffixes , prefixed by “Weg” to denote differential expression in Wnt3-MO early gastrulae , or given names based on recognizable repeats when present . The overall outcome of our in situ hybridization analyses was that transcripts identified as Wnt3-MO-underexpressed consistently showed Oral and Ingressing/Endodermal type expression profiles while the overexpressed ones all showed Aboral and Delayed type profiles . The significance level of the response did not , however , correlate with expression patterns ( O versus IE or A versus D , respectively; see z-scores in Table 1 ) . Remarkably , we were able in both cases to uncover a strong correlation when we included in the analysis the z-scores obtained for the Fz1-MO sample ( Figure 7A ) . This could be demonstrated by plotting the z-scores calculated for the two experimental conditions ( against non-injected ) against each other and determining the position of all the transcripts analyzed in Figures 2 and 3 , of genes with expression patterns characterized previously ( Bra , Fz3 ) and of five additional examples selected from our primary list ( FoxQ2c , Tbx; NotumO , sFRP-A , Gsc , WegO3; File S4; All patterns summarized in Table 1 and Figure 4 ) . Amongst the Wnt3-MO embryo under-expressed transcripts ( orange dots in Figure 7A ) , those with Fz1-MO z-values higher than -5 . 0 , ie not significantly affected or only relatively weakly underexpressed in Fz1-MO embryos , tended to show the O type expression pattern ( eleven of the thirteen examined transcripts in the dark orange “Class 1” zone ) . The others ( pale orange “Class 2” zone ) showed IE type expression profiles in eleven of the twelve cases . A similar strong correlation was found for the Wnt3-MO embryo-over-expressed transcripts ( green dots in Figure 7 ) . In this case , applying a Fz1-MO z-score value threshold of +5 . 0 we found that transcripts with higher z-values ( grey “Class 4” zone ) tended to show D or mixed D/A-type patterns ( seven and three respectively of the eleven analyzed transcripts ) , while nine transcripts with z-scores less than 5 . 0 ( green “Class 3 “zone ) showed A-type patterns and the tenth ( Notch ) a mixed A/D pattern . In this Class3 zone , responses to Fz1-MO were quite variable , including moderate over-expression , unchanged expression and , in a few cases , under-expression ( notably FoxQ2a and WegA1 ) . From these analyses we defined four “DGE classes” on the basis of z-score values in Wnt3-MO and Fz1 MO embryos , as indicated in Figure 7A . Although these classes strongly correlate with the four types of expression profiles ( Figure 7; Table 1 ) there are exceptions , for instance ZnfO is categorized as Class 2 on the basis of z-scores but shows an oral type expression profile , while Sulf is categorized as Class 1 but shows endodermal expression . Fz1 acts as a receptor for Wnt3 to activate Wnt/β-catenin signaling [39] , [40] , but is also thought to interact with the Clytia Strabismus protein to mediate planar cell polarity ( PCP ) , necessary for cell alignment in the ectoderm but also axial elongation during larval development and endoderm formation [41] . We thus hypothesized that the differences in expression responses in Fz1-MO versus Wnt3-MO could be due to the specific involvement of Fz1 in PCP . To test this hypothesis we made additional comparisons using a transcriptome derived from early gastrula embryos in which PCP was specifically disrupted by a morpholino targeting Strabismus ( Stbm-MO ) . Plotting the z-scores ( in relation to uninjected embryos ) of the Fz1-MO and Stbm-MO transcriptomes against each other revealed a striking similarity ( Figure 7B ) . The linear positive correlation was especially clear between Fz1-MO and Stbm-MO z-scores for the Wnt3-MO over-expressed transcripts ( i . e . DGE Classes 3 and 4; green and grey dots respectively in Figure 7B; Pearson correlation coefficient value 0 . 93 ) . The separation between Class 1 and Class 2 transcripts on the basis of Stbm-MO responses was less strict , with Class 1 transcripts showing moderately increased or decreased levels in these embryos , compared to unaffected or reduced levels in Fz1-MO embryos ( compare distribution of orange dots in Figure 7A and 7C ) . This can be explained by the requirement of Fz1 but not Stbm in Wnt/β-catenin signaling in the presumptive oral territory , We validated the transcriptome comparison analyses by in situ hybridization on Fz1-MO and Stbm-MO early gastrula embryos ( Figure 8 ) using a subset of the probes used to examine Wnt3-MO embryos ( Figure 5 ) . For each gene the expression patterns in the two morpholino conditions were strikingly similar: The Class 1/O-type pattern transcript Myb , and the Class 3/A-type pattern transcripts ZnfA and sFRP-A showed little change compared with non-injected controls ( Figure 8A , E , F ) . ZnfO , assigned to DGE Class2 despite its O-type expression profile , showed undetectable expression at the early gastrula stage in both Fz1-MO and Stbm-MO embryos ( Figure 8B ) and thus indeed represents an axially-expressed gene atypically sensitive to PCP perturbation . The weak change in levels of most axially-expressed genes along with the significant under-expression of FoxQ2a and WegA1 in both Fz1-MO and Stbm-MO early gastrula embryos ( Figure 7A , C; File S1 ) revealed in this study may at first seem difficult to reconcile with the previous description of an “aboralized” phenotype including a slight expansion of the FoxQ2a expression domain in Fz1-MO embryos [39] , but this can be explained by a difference in the timing of the two studies since the PCP effect is only transient . Thus , analysis of Stbm-MO embryos revealed that while aboral FoxQ2a expression is undetectable by in situ hybridization at the early gastrula stage it subsequently becomes restored , while conversely oral expression of Bra1 is transiently expanded but then becomes re-restricted to the oral pole of the planula [41] . The in situ analyses performed for Class 2/IE-type and Class4/D-type pattern transcripts also validated the DGE analyses . FoxA and Znf845 were barely detectable by in situ hybridization at the early gastrula stage ( Figure 8C , D ) , while Botch1 and bZip were detected strongly across the embryo ( Figure 8G , H ) . As in Wnt3-MO embryos ( Figure 5 ) the signal in these latter cases was mainly detected in cells positioned on the basal side of the ectodermal epithelial layer . We conclude that the relatively strong under-expression ( Class 2 ) or over-expression ( Class 4 ) of certain genes in Fz1-MO embryos is due in whole or part to disruption of PCP . This effect could reflect regulation of gene transcription by specific signaling pathways activated by PCP or be indirect , resulting from disturbed morphogenesis following failure of the ectodermal cells to align , to develop cell polarity and to undergo ciliogenesis [41] . To test whether the newly identified genes in Clytia were indeed involved with developmental processes as predicted by their expression patterns , we injected antisense morpholino oligonucleotides targeting a selection of identified genes . We included in this analysis transcripts representing each of the four expression profile types including cnidarian-restricted genes ( WegO1 , WegIE2 , WegD1 ) , candidate conserved developmental regulators ( Bra1 , Bra2 , FoxQ2c , FoxQ2a , HD02 ) and the partly conserved transcript WegA1 . For each morpholino tested , developmental defects observed at morphological ( Figure 9 ) and cellular ( File S8 ) levels were coherent with the corresponding expression patterns ( Figures 2 and 3 ) , confirming the usefulness of our approach to identify developmental regulators . Wherever possible ( 6/8 cases , see File S7 for details ) morpholinos targeting two different sites in the transcript were used , and in each case similar phenotypes were observed . Morpholinos to the three O-type expression pattern transcripts all showed defects in endoderm formation , consistent with endoderm fate specification in the oral territory [61] , [62] . Morpholinos targeting the two Clytia paralogs Bra1 and Bra2 both significantly inhibited endoderm formation . Initial signs of cell ingression at the oral pole occurred with only a slight delay with respect to non-injected controls , but subsequent filling of the blastocoel was strongly retarded , such that by 24hpf Bra1-MO and Bra2-MO embryos ( Figure 9C , D ) resembled uninjected embryos at the onset of gastrulation ( about 11hpf ) . Bra1-MO and Bra2-MO embryos then elongated somewhat and disorganized cells accumulated in the blastocoel to a variable degree , although often with a significant reduction in the amount of endoderm observed . Confocal microscopy confirmed that the residual ectodermal cells of both Bra1-MO1 and Bra2-MOe/i embryos accumulated in aboral regions and showed signs of epithelialization ( File S8 C , D ) . A similar but much less severe delay in gastrulation was obtained following injection of morpholinos targeting the cnidarian-restricted gene WegO1 , whose expression profile is very similar to that of Bra1 and Bra2 ( Figure 2 ) . Planulae showed a characteristic tapering of the oral half ( Figure 9H ) , and confocal microscopy revealed that endoderm was reduced in this region ( File S8 B ) . Strikingly , morpholinos targeting the A-type profile transcript WegA1 generated an opposite phenotype from the O-type pattern morpholinos . At the onset of gastrulation , massive cell ingression initiated widely across the embryo ( Figure 9F ) . This is reminiscent of the phenotype previously described for Fz3 MO [39] . During subsequent development , cells from the internal regions were expulsed in most embryos , so that by the planula stage , embryos were commonly smaller and consisted of accumulations of endodermal-type cells surrounded in some cases by a very thin ectoderm layer , in which the cells were stretched over the inner cell mass ( Figure 9F; File S8 G , H , I ) . Morpholinos targeting the two IE type pattern genes WegIE2 and FoxQ2c both caused only minor disruption of development prior to the end of gastrulation , but subsequent formation of the endodermal cell layer was affected , with in both cases a thin and uneven layer of endodermal cells observed at 48hpf surrounding a distended cavity containing cell debris ( Figure 9R , S ) . WegIE2-MO embryos showed additional disorganization of the oral ectoderm . Confocal microscopy confirmed that the endodermal cell layers were severely disorganized ( File S8 E , F ) . Finally , morpholinos targeting the two D-type profile genes , which are strongly up-regulated at the early gastrula stage upon Wnt3 , Fz1 or Stbm disruption , did not markedly disrupt gastrulation but resulted in highly aberrant morphology of the planulae ( Figure 9T , U ) . WegD1-MO embryos showed a distended aboral end with the ectoderm then becoming highly folded , this effect extending along the length of the embryo in the most extreme cases . Injection of morpholinos targeting the ANTP family gene HD02 also resulted in elongated and very irregular shaped planulae . In both cases the interface between the ectoderm and endoderm layers was very irregular with confocal microscopy revealing mixing of cells from the two layers and an absent or highly disrupted basal lamina between them ( File S8 P , T ) . In HD02-MO embryos , anti-tubulin staining revealed an abundance of neurite–like projections traversing irregularly this interface , contrasting with the well defined epithelial basal lamina and regular distribution of orthogonally extending neural projections in undisturbed planulae ( File S8; compare K and O ) . We used the strong correlation between DGE classes and expression patterns to assess the relationship between transcript identity and localization , using the 128 transcripts for which complete ORFs were present ( Figure 10 ) . The proportions of transcription factors and probable signaling pathway regulators were similar between DGE classes ( 12–21%; values not significantly different by Fisher's Exact Test ) . In contrast there was a significantly higher proportion of cnidarian-restricted sequences in DGE classes 1 , 2 and 3 than in DGE class 4 which tend to show D-type expression profiles ( around 30% vs 6%; Fisher's Exact Test p-value for this comparison = 0 . 04 ) . This analysis suggests that while cnidarian-restricted developmental regulators contribute significantly to patterning at the early gastrula stage , expression of evolutionary ancient genes predominates during development of the larva following gastrulation .
Our findings confirmed the central importance of Wnt signaling in embryo patterning . The transcripts under–represented in the spherical , aboralized Wnt3-MO embryos were during normal development systematically found expressed either in the oral ectoderm or in cells that contribute to the endodermal region ( defining O and IE type profiles respectively ) , while those from the over–represented set were detected either in the aboral ectoderm ( A type profile ) or generally repressed throughout the embryo at the early gastrula stage to be expressed in different patterns during planula larva formation ( D type profile ) . The O and A type profile genes displayed sustained localized expression at the poles through gastrulation and larval development and are thus good candidates for roles in patterning along the oral-aboral axis , but may also include precociously expressed gene markers of larval cell types enriched at one pole . We were intrigued to find that the four types of expression profile for Wnt3-MO-differentially expressed transcripts strongly correlated with four “DGE classes” , distinguished by the strength of the effect of Fz1-MO on the expression of the same genes . More specifically the axially expressed transcripts tended to show less extreme changes in expression in Fz1-MO early gastrulae than did IE and D-type profile transcripts ( Figure 7A ) . We have shown previously that Wnt/β-catenin signaling activated by Wnt3 and Fz1 is a key regulator of gene expression along the oral-aboral axis [39] , [40] . The relatively weak difference in expression of the axial genes in Fz1-MO relative to Wnt3-MO early gastrulae documented here could be explained , at least in part , by incomplete inhibition of this pathway by Fz1-MO compared with total extinction by Wnt3-MO , as revealed by β-catenin nuclear localization ( compare Figures 3 in [39] and [40] ) . It is also conceivable that Wnt receptors other than Frizzleds such as RYK or ROR2 [63] could be partly responsible for mediating the Wnt3 responses in oral regions . Our Stbm-MO analyses demonstrate , however , that the main explanation for less marked changes in expression of ‘axial’ versus ‘non-axial’ genes in Fz1-MO embryos relates to the involvement of Fz1 in PCP . One aspect of this is that transient up-regulation of some oral genes and down-regulation for some aboral genes due to PCP disruption , as shown in Stbm-MO embryos[41] , could in Fz-MO embryos counterbalance and dampen the effects of Wnt/β-catenin signaling . Concerning the non-axial genes the strong effects of PCP disruption could reflect direct signaling through ‘non-canonical’ intracellular pathways acting downstream of Fz/Dsh [64] . Given the transient nature of the effect , however , we favor the possibility that the effect is indirect , resulting from the developmental programs of the corresponding cell lineages being delayed or accelerated by a changed morphological environment . For cells of the presumptive endodermal region ( IE type pattern ) , lack of detection at the early gastrula stage in Fz1-MO and Stbm-MO embryos could result from disruption of ingression behavior due to loss of polarity of oral ectoderm cells . Conversely the strong over-expression of the D-type profile genes at the early gastrula stage in Fz1-MO and Stbm-MO embryos suggests that epithelial PCP may have a significant effect in delaying the development of certain planula cell types . One attractive possibility is that Fz-PCP disruption affects apical-basal polarity of epithelial cells and thus the generation of new cell types through oriented asymmetric divisions , as has been recently demonstrated in Xenopus embryos [65] . Consistent with this hypothesis , cells expressing Botch1 , bZip and Amt became prominent in basal regions of the epithelial ectoderm of early gastrulae when PCP was disrupted directly using Stbm-MO or Fz1-MO ( Figure 8 ) , or disturbed indirectly in the Wnt3-MO context [41] ( Figure 6 ) . Furthermore several other D type profile transcripts ( HD02 , UNC , WegD2 and possibly also Notch and Botch2 ) also tended to be expressed in basal regions of the ectodermal and/or endodermal epithelia during planula development ( Figure 3 ) . Our study provides further support for the well-known idea that a common set of transcription factors diversified from a common cnidarian-bilaterian ancestor has retained roles in regulating development in individual evolutionary lineages , with some families diversifying functions following lineage-specific gene duplications[4]–[6] , [9] . Clytia orthologs of many of known developmental regulator genes were identified from our unbiased screen based on sensitivity to Wnt/Fz signaling . All those tested showed characteristic spatiotemporally restricted expression profiles , and for four examples from well-known transcription factor families , roles in developmental regulation were supported by functional studies based on morpholino injection . Analysis of the morphant phenotypes suggested that the two Clytia Brachury paralogs Bra1 and Bra2 , expressed at the oral pole throughout larval development , both play important roles in controlling the progression of gastrulation . Expression around the blastopore has been proposed to be an ancestral metazoan characteristic of Brachury , which during bilaterian evolution became involved in the specification of various mesoderm and endoderm fates from these tissues [66] but with the ancestral role likely to have been in regulating morphogenetic movements [67] . In Clytia , although there is no blastopore , the relationship with the gastrulation initiation site is conserved , and our morpholino results suggest that morphogenetic movements upstream of endoderm specification are affected . The Hydra Bra1 and Bra2 orthologs have been shown to have subtly distinct roles in endoderm and ectoderm layers of the budding polyp [11] , suggesting that while embryogenesis roles for these genes overlap , their functions at other life cycle stages have diverged . A morpholino targeting FoxQ2c , expressed in the developing endodermal region during planula formation , caused severe defects in the organization of the endodermal layer . As with Brachyury , gene duplications have expanded the FoxQ2 gene family in Cnidaria , and in this case the paralogs have adopted clearly distinct expression profiles , FoxQ2a having conserved the likely ancestral aboral ( anti-blastoporal ) expression [68] while FoxQ2b is only expressed in oocytes [15] . The final member of a known developmental transcriptional regulator gene family we tested functionally was HD02 , a non-Hox member of the Antp homeodomain family [16] , expressed particularly strongly in cells at the base of the ectoderm and endoderm layers during larval development ( Figure 2P ) . The phenotypes following morpholino injections suggest that HD02 is involved directly or indirectly in regulating development of the neural network that develops at this site [69] , perhaps dependent on the correct organization of the basal lamina . Further in depth studies will be required to explore this possibility , as well as to confirm and understand fully all the other morpholino phenotypes documented here . On the basis of expression patterns it is likely that several other transcription factor genes identified in this study have developmental functions conserved through metazoan evolution . For example FoxA and FoxC are associated with distinct cell populations contributing to the endoderm region during gastrulation , as has also been reported for their Nematostella orthologs expressed in distinct regions of the developing pharynx [12] , [70] . In bilaterian species orthologs of these Fox genes are associated with development of endoderm/axial mesoderm and mesoderm respectively [71]–[74] . As well as transcription factors from families such as T-box , Fox and Antp , our transcriptome comparison identified likely regulators of a variety of intercellular signaling pathways including Notch , FGF , TGFβ and Ras-MAPkinase . These included core components ( ligands , receptors and secreted antagonists ) , but also less well known regulators acting in ligand or receptor processing and/or extracellular interactions , such as the Botch , Sulf and Notum proteins . Most strikingly we identified Clytia orthologs of known Wnt pathway regulators acting at all levels: Wnt ligands ( WntX1A ) , receptors ( Fz3 , Fz2 ) , members of three of the five families of secreted antagonists known from bilaterian models ( Dkk1/2/4; Dan1; two sFRPs ) [52] , [75] , MESD which specifically interferes with ligand co-receptor LRP5/6 [58] , [59] , an ortholog of the intracellular negative regulator Naked Cuticle [56] , and also the two Notum family lipases and Sulf . Sulf enzymes act on cell surface Heparan Sulphate Proteoglycans and have been reported to modulate Wnt as well as Hedgehog , TGFβ and FGF signaling while Notum releases the GPI anchor of glycipans such as Dally [76]–[79] . The oral expression profile of all the positive Wnt pathway regulators from this and our previous study ( five Wnt ligands , Axin and TCF ) reinforces the notion that an active Wnt signaling source is maintained at the cnidarian embryo and larval oral pole [40] , [42] , [80] as it is at the equivalent ‘head organizer” site in the Hydra polyp [81]–[83] . Co-expression of orally expressed putative pathway inhibitors such as Clytia NotumO is consistent with a role in limiting the extent of Wnt activity , equivalent to its action in Drosophila imaginal discs [76] or during planarian head regeneration [84] . Most of the putative Wnt antagonists we identified , however , were expressed aborally in the gastrula and in aboral pole subdomains in the planula ( demonstrated by in situ hybridization for Dkk1/2/4 , Dan1 , sFRP-A and NotumA , implied by DGE responses for sFRP-B and MESD ) , suggesting that Wnt signaling is inhibited actively at the aboral pole region in the larva . Future functional studies will be required to examine the functions of each Wnt regulator during Clytia development , and to unravel the interactions between them . Our study uncovered many potential developmental regulators amongst gene families with orthologs and/or shared domains identifiable from the mass of available genomic and transcriptomic data across bilaterian species , but for which nothing is known about function or expression . These include zinc finger and helix-loop-helix domain transcription factors as well as putative novel signaling pathway components . The prominence of cell surface protein modifiers with known impact on one or several signaling pathways in our screen raises the possibility that some of the other uncharacterized conserved or cell surface proteins may function similarly . In this context it would be interesting , for example , to test the function of the ZpdA and Aat genes , which code for a likely cell surface glycoprotein and a membrane transport protein respectively . Uncovering developmental roles for such proteins in Clytia would open the way to explore the involvement of potential novel regulators of key embryonic and cellular processes in bilaterians , and the associated evolutionary and medical implications . WegA1 offers an interesting illustration of this possibility . WegA1-MO injection results in a spectacular developmental defect involving premature cell ingression ( a process of epithelial-mesenchymal transition ) at gastrulation , and a massive shift in the balance of ectoderm to endoderm formation . This finding implies that this previously unknown protein functions during normal development under the control of Wnt/β-catenin and PCP signaling to inhibit cell ingression in aboral territories . As well as the 135-amino acid , C-terminal DUF3504 domain the WegA1 sequence contains a putative nuclear localization signal . Whether it has true orthologs in bilaterians remains to be established . Amongst the potential developmental regulators identified in our study , 29% were defined as cnidarian-restricted on the basis that they had no identifiable orthologs in any other metazoans . Previous surveys of available cnidarian genomic and transcriptomic data revealed about 25% in Clytia and 15% in the ‘polyp only’ cnidarian models Nematostella and Hydra [4] , [14] , [18] , [23] . A few of these match genes previously known only outside Metazoa , and so represent ancient genes lost in bilaterian branches or gained by lateral gene transfer , while the others probably represent cnidarian innovations . Although more in depth studies of each gene are required , the characteristic phenotypes observed in our morpholino experiments support the stereotypical expression pattern data in suggesting roles in regulating developmental processes for these cnidarian-restricted genes: larval oral pole organisation for WegO1 , endoderm formation for WegIE2 and epithelial organization for WegD1 respectively . More than half of the cnidarian-restricted transcripts identified in our study contained secretion signal sequences . These are prime candidates for roles in cell-cell signaling , either as ligands or as modulators of ligand/cell surface/receptor interactions during axis establishment and gastrulation . Candidate receptors for such signaling molecules include the many unclassified 7tm receptors identified particularly amongst IE profile/DGE class 2 transcripts . With notable exceptions such as Frizzled and Patched , members of the 7tm superfamily , including the G–protein coupled receptors ( GPCRs ) , have not been strongly implicated in developmental regulation in bilaterians . This family has expanded independently in cnidarians [14] , so its exploitation for developmental signaling might represent a cnidarian specialty , a fascinating possibility to explore in future studies . Intriguingly , almost all ( 35/37 ) of the cnidarian-restricted genes we identified belonged to the three DGE classes associated with regional expression and thus embryo patterning at the gastrula stage ( Figure 10 ) . Conversely , the DGE Class 4 transcript set contained a higher proportion of broadly conserved “ancient” genes . Recent studies have demonstrated that the extensive variation in modes of early embryogenesis between species correlates with expression of evolutionarily “newer” genes , while subsequent ‘phylotypic stages’ ( corresponding to neurula and somatogenic stages in vertebrates and the germ-band segmentation stage in insects ) are strongly conserved at the phylum level and tend to express more ancient genes [85] , [86] . With the caveat that our analysis concerns only a small fraction of the transcriptome and provides only limited coverage of developmental stages , the observation that most ( 28/35 ) of the DGE class 1–3 ( putative patterning ) genes lacked counterparts in Nematostella or Hydra may reflect the widely divergent modes of early embryo patterning and gastrulation amongst cnidarian species [87] . In contrast several of the DGE class 4 genes , mostly “ancient” , appeared to be associated with epithelia development and in particular with formation of the basal lamina , a structure considered to be a major innovation in the animal lineage [88] , [89] and highly conserved in all Eumetazoa . A temporal shift in expression from “new” to “old” genes between gastrula and larva in cnidarian species is consistent with the idea that the epitheliarized , planula stage-ciliated torpedo larva represents the phylotypic stage [90] . To conclude , from a methodological standpoint , our study demonstrates the power of rigorous unbiased transcriptomic approaches to obtain a fresh view of gene conservation and innovation in the evolution of animal diversity . It also illustrates how transcriptome comparisons can allow prediction of expression characteristics without doing large-scale in situ hybridization screens; The differential transcriptional responses in Fz1-MO and Stbm-MO embryos will be very useful for picking candidate genes for future studies targeted to particular developmental processes . From a theoretical standpoint , our findings provide strong support for the notion that many evolutionary-conserved genes are deployed across eumetazoans to regulate development , but also good evidence that developmental regulation in cnidarians may involve a significant number of taxon-restricted genes . Functional studies of the genes identified here in Clytia should provide a fruitful entry for exploring both these possibilities .
Eggs obtained by light-induced spawning of laboratory-raised medusae were microinjected with morpholino oligonucleotides prior to fertilization as described [39] . Previously unpublished morpholino sequences are provided in File S7 . Use of genetically identical female medusae derived from a single individual laboratory polyp colony Z4B and males from a closely related colony [32] restricted the problems of sequence polymorphism . After culture at 18°C to the four cell stage , any unfertilized or abnormally-dividing embryos were removed . Early gastrula stage embryos , used for RNA extraction or fixed for in situ hybridization or confocal microscopy , were obtained after culture at 16°C overnight ( 17 hours ) . Planulae were fixed for in situ hybridization after 24 or 48 hours of culture at 18°C . Particular care was taken to use identical timing and temperature regimes for all experiments . For each experimental condition , total RNA was extracted from batches of 900–1400 early gastrula stage embryos using RNAqueous kit ( Life Technologies/Ambion , CA ) . RNA integrity was confirmed by formaldehyde gel electrophoresis . Two independent biological replicates were performed for the uninjected and Wnt3-MO conditions , and single samples for the other morpholino conditions . Estimated final embryo numbers in each sample , after removal of any showing arrested cleavage or irregular development , were as follows: Uninjected: each 1900; Wnt3-MO: each 2300 , Fz1-MO: 900 , Fz3-MO: 1600 and Stbm: 1400 . Library construction and Illumina short-read ( 51 bp ) sequencing was performed by GATC ( Konstanz , Germany ) . To quantify gene expression , the number of mapped reads onto a reference transcriptome data set was taken as a measure of transcript level . The reference transcriptome , comprising 24893 distinct ( non-overlapping ) assembled sequences , was built by combining , using CAP3 software , previous EST data [15] , [32] and Illumina sequences from one of the untreated early gastrula samples generated in this study . Redundant sequence entries were eliminated by USEARCH ( ver . 5 . 2 . 32_i86linux32 ) . The longest predicted ORF from each sequence was used as the reference for read mapping . To reduce polymorphism , adaptator sequences and probable 5′ UTR sequences upstream of the first ATG in each cDNA contig were removed , For each experimental condition approximately 80 million of 51pb Illumina reads were mapped on the reference transcriptome using the Bowtie command , with tolerance of two mismatches . Reads that matched to more than one reference sequence were not taken into account . Around 35% of the reads obtained for each condition could be mapped using this method . Statistical analysis was performed using the DEGseq R package [50] to determine for each transcript whether the observed ratio of transcript levels ( M ) between two samples is significant given the global average expression ( A ) . The Random Sampling Model employed assumes a normal distribution for log2 ( C ) , where C is the number of counts , as confirmed for our data by a Q-Q plot ( Figure 1B ) . M = log2 ( C sample1 ) -log2 ( C sample2 ) estimates the difference of expression between the conditions; A = ( log2 ( C sample1 ) +log2 ( C sample2 ) ) /2 measures the average expression in the two conditions . A p-value was generated for each gene to determine whether the expression difference between samples was significant . A z-score was generated for each transcript , as a measure of the deviation from the random model ( z-score = ( Mobserved – Mexpected according to random sampling model ) /Var ( M expected according to random model ) . The MATR method used an estimation of the variation between duplicate embryo samples ( calculated using the CTR method ) to generate a second MA plot and to adjusts the z-score accordingly . A R-script was devised to analyze automatically the six possible reading frames of each unique assembled transcript sequence and to predict the best ORF ( “find . ORF” script downloadable at http://octopus . obs-vlfr . fr/R_scripts ) . Sequence comparisons were performed with both BLASTx with the whole sequence and BLASTp with predicted translated ORF against the “non-redundant’” ( nr ) NCBI database . Domain analyses ( Files S1 and S3 ) were performed using Interproscan , SignalP for the detection of secreted peptide signals and TMHMM for the prediction of transmembrane domains . Gene identities ( column 3 of File S1 ) were based on BLAST and domain analyses . Gene accession numbers are provided in File S1 and File S3 . To determine orthology of the transcript sequences studied in detail ( Table 1 ) we searched for homologs by reciprocal BLASTp . When reciprocal blast and domain analysis ( see above ) gave unambiguous identities ( non-multigene families ) , gene names were attributed directly ( Sulf , Aat , Asparaginase , Amt , UCP ) . For certain multigenic developmental regulator families , we added our candidate sequence and the retrieved cnidarian sequences to alignments from previously published studies kindly provided by authors ( see File S2 and acknowledgements ) . Where no existing appropriate alignments were available , sequences from a range of eumetazoan genomes ( Drosophila melanogaster , Lottia giganta , Strongylocentrotus purpuratus , Xenopus laevis or Homo sapiens , H . magnipapillata and N . vectensis ) were aligned using MUSCLE , the best fitting model of evolution was determined using ProtTest2 . 4 , and phylogenetic analysis performed using PhyML3 . 0 . The trees are available in File S2 . In cases where clear Hydra magnipapillata orthologs were identified , further analysis was performed using the Hydra vulgaris transcript dataset ( HAEP ) available at http://compagen . zoologie . uni-kiel . de/blast . html ) [91] . The number of matching reads recorded in each separated cell population ( endoderm , ectoderm and nanos-positive cells ) was normalized with respect to total read number ( File S5 ) . For cnidarian-specific sequences , WegO1 , WegO2 , WegIE2 , WegA2 , WegD2 , Zpd , had no recognisable homologs in Hydra or in Nematostella genomes . For WegIE1 , WegD1 we identified single orthologs in Hydra: ( listed in File S5 ) . In situ hybridization probes were synthesized from cDNA clones corresponding to our EST collection when available . For the remaining sequences , cDNAs were cloned by PCR using the TOPO-TA cloning kit ( Invitrogen ) . All sequences were verified before probe synthesis . DIG-labeled antisense RNA probes for in situ hybridization were synthesized using Promega T3/T7/Sp6 RNA polymerases , purified using ProbeQuant G-50 Micro Columns ( GE Healthcare ) and taken up in 100 µl of 50% formamide . Gastrulae , 24hpf and 48hpf planula larvae were fixed in 3 . 7% formaldehyde/0 . 2% glutaraldehyde in PBS for 2 hours on ice , washed five times in PBST ( PBS containing 0 . 1% Tween 20 ) for 10 minutes , dehydrated in PBST/50% methanol and stored in methanol at −20°C . In situ hybridization was performed using the InsituPro robot ( Intavis ) . After rehydratation in PBST/50% methanol and three 5 minute washes in PBST , samples were transferred to the plate . The robot program was as follows: two 20 min washes in PBST; 20 min in PBST/50% hybridization buffer ( 5X SSC , 50% deionized formamide , 1% dextran sulfate , 1% SDS , 50 µg/ml tRNA , 50 µg/ml heparin ) ; 20 min in hybridization buffer; 2 hours pre-hybridization in hybridization buffer at 62°C; 40 to 63 hours hybridization at 62°C with the denatured DIG-labelled RNA probe; four 30 min washes in 5X SSC , 0 . 1% Tween 20 and 50% formamide at 62°C; four 30 min washes in 2X SSC , 0 . 1% Tween 20 and 50% formamide at 62°C; two 20 min washes in 2X SSC , 0 . 1% Tween 20 at 62°C; two 20 min equilibration steps in MABT ( 100 mM maleic acid pH 7 . 5 , 150 mM NaCl , 0 . 1% Triton X-100 ) ; 1 hour blocking in MABT/1% blocking reagent ( Roche ) ; 3 hours incubation with an alkaline phosphatase labeled anti-DIG antibody diluted 1/2000 in the blocking solution; seven 20 min washes in MABT; three 20 min washes in TMNT ( 100 mM Tris-HCl pH 9 . 4 , 50 mM MgCl2 , 100 mM NaCl and 0 . 1% Tween 20 ) . The color reaction was performed manually in TMNT containing 0 . 08 mg/ml NBT and 0 . 1 mg/ml BCIP ( Promega ) . Color development time varied from 1 hour to 1 day . Samples were then washed twice in water , three times in PBS , post-fixed in PBS/3 . 7% formaldehyde and washed three times with PBST before mounting in 40% glycerol . For the selected candidate genes we addressed phenotype specificity by designing and testing several morpholinos targeting different parts of the sequence , discarding any that proved toxic to cell division during pre-gastrula development . We could only identify 1 non-toxic morpholino targeting FoxQ2c and WegA1 , and none for FoxQ2a . For Bra2 one morpholino targeted the predicted AUG translation initiation codon and the other an exon-intron junction ( all details in File S7 ) . For each morpholino we first injected a range of concentrations into eggs prior to fertilization , and then assessed planula morphology for the lowest non-toxic dose at 24 h and 48 h . The cellular basis of the observed phenotypes was then further assessed by confocal microscopy . Images of in situ hybridization profiles and DIC images of live embryos were acquired on an Olympus BX51 microscope . For confocal imaging of cell boundaries using fluorescent phalloidins and nuclei using Hoechst 33358 or TOPRO-3 dyes , embryos were fixed , processed and imaged on a Leica SP5 microscope as described previously [39] . Microtubules were stained by immunofluorescence using anti-α tubulin rat monoclonal antibody YL1/2 ( Sigma ) followed by rhodamine-conjugated anti-rat Ig antibodies ( Jackson Immunoresearch ) . Total RNA from 60 Wnt3-MO injected and 60 non-injected early gastrulae was extracted using RNAqueous-Micro kit according to the manufacturer's instructions ( Ambion , Warrington , UK ) . Genomic DNA was removed by a DNAse I treatment ( Ambion ) and this step was controlled for each RNA extract . First-strand cDNA was synthesized using 500 ng of total RNA , Random Hexamer Primers and Transcriptor Reverse Transcriptase ( Roche Applied Science , Indianapolis , USA ) . Quantitative PCRs were run in quadruplicate and EF-1alpha used as the reference control gene . Each PCR contained 5 µl cDNA 1/400 , 10 µl SYBR Green I Master Mix ( Roche Applied Science ) , and 200 nM of each gene-specific primer , in a 20 µl final volume . PCR reactions were run in 96-well plates , in a LightCycler 480 ( Roche Applied Science ) . Sequences of forward and reverse primers designed for each gene: EF-1alpha-F 5′ TGCTGTTGTCCCAATCTCTG 3′; EF-1alpha-R 5′ AAGACGGAGTGGTTTGGATG 3′; Bra-F 5′ GCAACACCCACAACAACAAC 3′; Bra-R 5′ TACGGGAAACATACGCCTTC 3′; NotumO-F 5′ GGGACATCTAAAACCCATGC 3′; NotumO-R 5′ CATGGATCTCGCATTGTGAC 3′; ZnfO-F 5′ TGCTGCTAACAACGACCAAC 3′; ZnfO-R 5′ TGGTGGAAGTGGAGATTGTG 3′; Mos3-F 5′ ATCTTACGTCCCGAACAACG 3′; mos3-R 5′ ATCCACCAATGGCAGCTTAC 3′; Znf845-F 5′ AGACCGACAGCATTTCATCC 3′; Znf845-R 5′ TGGCATTCCTTGCATACCTC 3′; Dkk1/2/4-F 5′ GGGCTTGTTCGTACTTTTCC 3′; Dkk1/2/4-R 5′ ATTCCATCCCACGACAACAC 3′; ZnfA-F 5′ CAACAACTTTCACCGAGCTG 3′; ZnfA-R 5′ TGTCTCTTGTGTTGCCAAGC 3′; Dan1-F 5′ CATGCCCGTTCATGAGAAAG 3′; Dan1-R 5′TTTTGGCTGTTCCCACTGTC 3′; NotumA-F 5′ TGCTGAAGGGTCGTACATTG 3′; NotumA-R 5′ CGTGTGTCCATTTTCAGTGC 3′; HD02-F 5′ TT AACAGCCCACCGAAACTC 3′; HD02-R 5′ CGTCGTGTTTTTCAGTGACG 3′ . For each gene studied an expression level N was calculated as 2−Ct , where Ct ( Cycle threshold ) represents the number of cycles required for the fluorescent signal to cross the threshold . | The recent wave of genome sequencing from many species has revealed that most of the gene families known to regulate animal development are shared not only between humans and laboratory favorites such as mice , flies and worms , but also by evolutionarily more distant animals such as jellyfish and sponges . It is often assumed that genes inherited from a common ancestor remain largely responsible for regulating embryogenesis across these animal species , rather than more recently evolved genes . To address this issue we made an unbiased , systematic search for developmental genes in embryos of the jellyfish Clytia , selecting genes whose expression altered upon manipulation of the key regulator Wnt3 , and comparing their expression in embryos specifically disrupted for Planar Cell Polarity . Identification of evolutionarily conserved and novel genes as developmental regulators was confirmed by demonstrating characteristic expression profiles for a sub-set of genes , and by gene knockdown studies . Conserved genes coded for members of many known signaling pathway and transcription factor families , as well as previously unstudied proteins . Nearly 30% of the identified genes were restricted to cnidarians ( the jellyfish-sea anemone-coral group ) , supporting the idea that the appearance of new genes during evolution contributed significantly to generating animal diversity . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"developmental",
"biology",
"cell",
"biology",
"biology",
"and",
"life",
"sciences",
"computational",
"biology",
"evolutionary",
"biology"
] | 2014 | Differential Responses to Wnt and PCP Disruption Predict Expression and Developmental Function of Conserved and Novel Genes in a Cnidarian |
Hard-wired , Pavlovian , responses elicited by predictions of rewards and punishments exert significant benevolent and malevolent influences over instrumentally-appropriate actions . These influences come in two main groups , defined along anatomical , pharmacological , behavioural and functional lines . Investigations of the influences have so far concentrated on the groups as a whole; here we take the critical step of looking inside each group , using a detailed reinforcement learning model to distinguish effects to do with value , specific actions , and general activation or inhibition . We show a high degree of sophistication in Pavlovian influences , with appetitive Pavlovian stimuli specifically promoting approach and inhibiting withdrawal , and aversive Pavlovian stimuli promoting withdrawal and inhibiting approach . These influences account for differences in the instrumental performance of approach and withdrawal behaviours . Finally , although losses are as informative as gains , we find that subjects neglect losses in their instrumental learning . Our findings argue for a view of the Pavlovian system as a constraint or prior , facilitating learning by alleviating computational costs that come with increased flexibility .
The functional architecture of responding involves two fundamental components that are behaviourally [1] and computationally [2] separable: Pavlovian and instrumental . The instrumental component respects the stimulus-dependent contingency between responses and their outcomes ( stimulus-response and action-outcome learning ) [3] . By contrast , preparatory Pavlovian responses , chiefly involving approach and withdrawal , are elicited by the appetitive or aversive valence associated with predictive stimuli in a manner that is not dependent on the consequences of those responses [3]–[5] . The interactions between the two systems are most evident when automatically-elicited Pavlovian responses interfere with contingent instrumental responding [1] , [6]–[9] . For instance , pigeons will strikingly continue to peck at a light predictive of food ( a preparatory approach elicited by the appetitive prediction ) , even if the food is withheld every time they peck the light ( the instrumental contingency ) [10] , [11] . Pavlovian interference likely contributes to many quirks of behaviour such as impulsivity [12] , framing and [13] , endowment effects [14] and many other “anomalies” [15] , including neurological [16]–[19] and psychiatric diseases [20]–[26] . Further , puzzling facets of seemingly purely instrumental behaviour such as the difficulties in learning ‘go’ responses to avoid punishments; or ‘nogo’ to obtain rewards ( unpublished data ) and even the restrictions in associations evident in ‘evolutionarily preparedness’ [27] , [28] might be traced to Pavlovian principles . However , instrumental and Pavlovian systems share overlapping neural hardware . Their bidirectional interaction is characterised by two key triads: rewards are tied to approach and vigour; and punishments to withdrawal and behavioural inhibition . The neuromodulator dopamine ( DA ) responds predominantly to rewards [22] , [29]–[31] , induces behavioural activation and enhances approach [32]–[35] . Each aspect of this triad confounds the role of the phasic DA bursts in the flexible acquisition of instrumental values [36]–[42] . Serotonin appears to lie at the heart of the aversive triad , having been linked to punishments [43]–[45] , behavioural inhibition and withdrawal [25] , [32] , [46]–[52] , although dopamine acting via D2 receptors likely also plays a role in linking absence of rewards to nogo [17] , [53] , [54] . Signatures of both triads are also evident in neural circuits involved in response and choice . In the dorsal striatum , there are interdigitated pathways for ‘go’ and ‘nogo’ , with the go pathways again linked positively to rewards via dopamine [16] , [18] , [55] , [56] . The ventral striatum is primarily organized along an appetitive/aversive axis with direct links to approach and withdrawal behaviours [57] , [58] . The aversive triad is also tightly linked to the dorsal raphé and the periaquaeductal gray [59] , [60] . The main routes to the scientific investigation of these interactions consists of tasks in which Pavlovian stimuli are presented during ongoing instrumental tasks . However , these have as yet not explored the full set of interactions characterising the overlap between the two systems . Two critical confounds remain: The first confound concerns the precise nature of the effect of Pavlovian stimuli on instrumental behaviours . The instrumental behaviours studied have largely been appetitively motivated approach behaviours ( in Pavlovian-Instrumental Transfer ( PIT ) and conditioned suppression tasks , [1] , [6]–[8] , [61]–[63] ) , and one instance of aversively motivated withdrawal behaviour [64] . The relative role of the appetitive-aversive motivation axis versus that of the approach-withdrawal axis is unknown . This in turn obscures the nature of the interaction: whether Pavlovian stimuli interact with the value of the instrumental behaviour , or by promoting specific responses [1] , or even simply by modulating behavioural activation [5] . Second , the extent to which the separation of reward and punishment processing into opponent motivational structures applies to instrumental as well as Pavlovian learning is incompletely explored [1] , [27] , [28] , [65] . All these issues can simultaneously be addressed in a combined PIT and conditioned suppression task with both approach and withdrawal actions in which the overall motivational component of approach and withdrawal are matched ( Figure 1 and Table 1 ) . The task separates the contributions of approach and withdrawal by using two counterbalanced blocks , one involving approach go versus nogo , and the other withdrawal go versus nogo . The comparison between go and nogo controls for effects of behavioural activation or inhibition . In each block , subjects first underwent brief instrumental training ( Figure 1A ) , learning from positive and negative feedback ( monetary gains and losses of €0 . 20 ) whether to produce a go or a nogo response associated with sorting mushrooms . In the approach block ( Figure 1A , top , all 46 subjects ) , go responses involved moving the cursor onto a mushroom ( to collect it ) , while nogo involved doing nothing , thus not collecting the mushroom . To test for the effect of low-level motor variables , subjects performed one of two types of withdrawal actions . In “throwaway” ( 24 subjects , Figure 1A , middle ) , go involved moving the cursor physically away from the mushroom and clicking into an empty blue box; nogo involved doing nothing , and thus keeping the mushroom . Importantly , both approach to and withdrawal from the instrumental stimulus were orthogonal to any approach and withdrawal that might be directed at the Pavlovian background stimulus . In “release” ( 22 subjects , Figure 1A , bottom ) , the subjects had to start by pressing the mouse button . Go involved releasing the button to avoid collecting the mushroom; nogo involved continuing to press the button and thereby receiving the mushroom . In order to orthogonalise the approach-withdrawal and appetitive-aversive axes , the learned instrumental values in approach and withdrawal blocks needed to be matched . To achieve this , both go and nogo responses were , if correct , rewarded . Additionally , to avoid the confound of activation , in each block ( i . e . in both approach and withdrawal blocks ) the go action was designated as the correct response to half the instrumental stimuli , and the nogo action to the other half ( see Table 1 ) . Incorrect responses had opposite outcome contingencies to correct responses , yielding more punishments than rewards . This ensured that go , nogo , approach and withdrawal overall had the same learned association with rewards and punishments . We tested both deterministic and probabilistic outcomes but found no differences . In the second part of each block , subjects passively viewed unrelated , fractal , stimuli paired with separate auditory tones ( Figure 1B ) . Each compound Pavlovian stimulus was deterministically associated with a monetary gain or loss , i . e . its Pavlovian value was equal to that monetary outcome . Every fifth trial in the Pavlovian block was a query trial ( Figure 1C ) , in which subjects chose the better of two fractal visual stimuli without being informed about the outcome . Finally , in the PIT stage , the instrumental stimuli were presented on a background of fractal Pavlovian stimuli together with the auditory tones , and again without outcome information . Our task addressed the key confounds described above . With respect to the triads , we found that the Pavlovian influence is action specific: appetitive Pavlovian cues boosted go approach responses and suppressed withdrawal go responses; aversive Pavlovian cues did the opposite . Additionally , subjects were substantially biased against withdrawal , but we found no evidence that the instrumental learning component itself differed between the approach and withdrawal condition .
There was no difference between the results for probabilistic and deterministic feedback , and we therefore present the combined data . Analysis of the components of the experiment indicate robust , yet moderate , instrumental conditioning that was stable during the PIT period , combined with highly robust Pavlovian conditioning . Figure 2A shows the instrumental probability of choosing the more rewarded ( “correct” ) stimulus over time . Subjects rapidly came to prefer the more rewarded action . Preference was weaker for go withdrawal , against which there was a consistent bias . We intended the instrumental preference to be weak to avoid ceiling effects when assessing PIT . Subjects also exhibited predictable variability on a shorter time-scale: Figure 2B shows the immediate consequences of rewards and punishments on subsequent behaviour . It is notable that punishments did not reduce the repeat probability below chance level ( mean is not , one-tailed t-test ) . The same was found when analysing go and nogo choices separately: in both cases , was not significantly different from 0 . 5 ( both , two-tailed t-test ) , and was significantly smaller than ( both , paired t-test ) . Whether this really does represent an insensitivity to punishments depends , however , on the average stay probability , and on how this average stay probability is related to past reinforcements . Subjects were instructed that the outcomes of responses in the PIT block would be counted as in the instrumental block . Figure 2C shows that this led to stable maintenance of the instrumental response tendencies throughout the PIT block . Figure 2D shows that all but one ( excluded ) subject showed extremely good performance on the Pavlovian query trials interleaved with the Pavlovian training ( mean correct ) . Given the success of instrumental and Pavlovian training , we next analysed the raw effect of Pavlovian stimuli on approach and withdrawal choices . Figure 2E shows a highly significant interaction between block and Pavlovian stimulus valence . Relative to neutral stimuli , positive Pavlovian stimuli enhanced approach and inhibited withdrawal go over nogo . Conversely , negative Pavlovian stimuli enhanced withdrawal and inhibited approach go over nogo . A similar analysis looking at the probability of responding incorrectly ( outside the blue box ) showed no effect of the Pavlovian stimuli in either approach or withdrawal condition and no interaction ( respectively , ANOVA ) , suggesting that these results were not due to response competition . Note that the withdrawal go probabilities were lower than the approach ones , again reflecting the overall bias against go withdrawal . Average reaction times for go approach and go withdrawal actions did not differ ( , 2-tailed t-test ) . Against our expectations , Pavlovian stimuli of both positive and negative valence shortened reaction times in a parametric manner relative to neutral Pavlovian stimuli ( Figure 2F , p = 0 . 0310 , ANOVA ) , although this effect was not present in either block separately ( p = 0 . 5502 and p = 0 . 0781 respectively , ANOVA ) . The size of the PIT effect may have been affected by the extent of instrumental learning ( and thus the actual learned action values ) , by response biases , and by generalization from the instrumental to the PIT stage . In addition , there may have been differences in the instrumental learning of approach and withdrawal actions ( Figure 2A ) . We decomposed and analysed all such factors using a detailed reinforcement learning model . This contained explicit parameters capturing all the instrumental and Pavlovian effects in the task , and was fit to the choice data of all subjects . We used group-level Bayesian model comparison [66] to choose amongst a variety of model formulations ( reporting scores relative to the final model ) , and ensured that inference yielded correct parameter estimates when run on surrogate data generated from the assumed underlying decision process . The final model included parameters associated directly with the instrumental requirements of the task . These comprise one learning rate ; two parameters and representing the bias towards go in the approach and withdrawal blocks; and two separate free parameters and , representing the effective strengths of rewards and punishments . At a group level , subjects were biased against active withdrawal , but showed no bias for or against approach ( and respectively , two-tailed t-test ) , the difference being significant ( , ANOVA , Figure 4A ) . Withdrawal biases in the release and throw away experimental subgroups did not differ ( , ANOVA ) , controlling for motor effects . The withdrawal bias accounts for the lower performance on go withdrawal in Figure 2A . One concern is that differences in the biases might have masked differences in learning ( i . e . the reward sensitivities ) in the approach and withdrawal conditions . We tested this by allowing for separate reward and punishment sensitivities in the two conditions ( Model 6 ) or separate learning rates ( Model 7 ) . The use of these extra parameters was structurally rejected by the model selection process ( respectively for the purely instrumental trials ) ; and the freedom to choose different parameter values in these conditions was duly not used ( Figure 5 ) . The absence of any difference in the learning parameters for approach and withdrawal suggests that the instrumental system treated approach and withdrawal entirely equally . We will see below that this was not true for the Pavlovian system . Although , by design , rewards and punishments were equally informative , subjects chose to rely more on rewards than punishments ( Figure 4B ) . Rewards had a stronger effect than punishments both at a group level and for all individual subjects , the difference being significant ( , ANOVA ) . Indeed , the average punishment sensitivity was not distinguishable from zero ( , two-tailed t-test ) . This remained true when we separately tested subjects who were given deterministic ( , two-tailed t-test ) and probabilistic ( , two-tailed t-test ) feedback . Supplementary analyses ( Text S1 ) excluded two further explanations for the punishment insensitivity: first , that it is due to choice perseverance ( Figure S1 Text S1 ) ; and second that it is due to an emerging maximisation behaviour ( Figure S2 in Text S1 ) . Thus , it appears that the pattern seen in Figure 2B is indeed due to a differential sensitivity to rewards and punishments . We next analysed the generalization of instrumental values from the instrumental to the PIT blocks . Generalization could be imperfect in two ways - the starting values in the PIT block could differ from the ending values in the preceding instrumental block , and the values could then decay over time or trials during the PIT block given the lack of information about the outcomes . We constructed models including such effects , and tested whether their excess complexity was outweighed by their fit to the data . As expected from the stable raw probabilities of choosing the correct ( i . e . , more rewarded ) option ( Figure 2C ) , a model in which the instrumental values decayed exponentially over time during the PIT block ( mimicking extinction ) did not provide a good account of the data ( Model 9 , compared to Model 10 ) . Rather , the final model allowed for the addition of random generalization noise to each . These factors were drawn independently from the same normal distribution for all stimulus-action pairs , and the mean and variance of this distribution were both inferred without constraints ( see Methods ) . Figure 4C visualizes the resulting changes; each dot represents the preference for the go action ( ) for all subjects and all stimuli . The abscissa shows this at the end of the instrumental stage , the ordinate after addition of the noise for the PIT stage . Importantly , there was no systematic difference in mean correct action values either in the instrumental or PIT stage ( Figure 4D ) . We were mainly interested in the effect of the Pavlovian values on instrumental performance . We therefore fitted unconstrained parameters to separately capture the influence of each of the five Pavlovian stimuli on instrumental go actions in both the approach and withdrawal condition . All models accounted for performance in the PIT part by adding up instrumental and Pavlovian influences prior to taking a softmax [67] , [68] . This amounts to treating instrumental and the Pavlovian controllers as separate experts , each of which ‘voted’ for its preferred action . The model captured in detail , and thereby controlled for , variability in instrumental learning and generalization . The final model predicted the choices of every individual subject better than chance ( binomial probability , for every subject , overall predictive probability 0 . 7544 ) . The maximum a posteriori ( MAP ) estimates of this model's parameters painted a picture very similar to that seen in the raw data . Figure 4E shows the parameters of the model related to the influence of each Pavlovian stimulus . The pattern mirrored that seen in the raw data: there are highly significant , and opposite , effects in the approach and withdrawal blocks , with appetitive stimuli ( ++ and + ) promoting approach but inhibiting withdrawal; and aversive stimuli ( -- and - ) promoting withdrawal but inhibiting approach . At a single subject level , the effect in the approach block was seen in 45/46 subjects ( 98% ) , while it was seen in 30 subjects ( 65% ) in the withdrawal block . Since there was no difference in the learned value of go or nogo actions in either approach or withdrawal blocks , and in either the instrumental learning or the PIT stages ( Figure 4D ) , any PIT effects are unlikely to be due to a preferential association of a Pavlovian stimulus with the learned value of an action . Rather , they reflect the approach or a withdrawal nature of the action . We included two separate groups of subjects who either performed a throwaway withdrawal action , or a release withdrawal action . This was both to test the contribution of an approach/withdrawal component aimed at the Pavlovian stimuli tiling the background , and in recognition of the sophistication of defensive reactions [27] . Figure 4F shows that Pavlovian stimulus value had a significant , linear effect on both withdrawal action types , and that this overall linear effect did not differ between the two action types . At an individual level , linear correlations were positive for 16 ( 72% ) and 14 ( 58% ) subject in the release and throwaway condition , respectively . No psychometric measure of anxiety or depression correlated with any of the parameters in the main model .
One of the central motivations for our investigation was the observation that the neural substrate does not respect the logical independence of reward/punishment and approach/withdrawal . Rather , as we have discussed , these are tied together , via the structure of the striatum and also specific neuromodulators . While the neural basis for the promotion of approach responses by appetitive stimuli is known to involve both amygdala and striatum [62] , [63] , [75] , the neural bases for the effects of aversive Pavlovian stimuli are less clear . There are no data on withdrawal responses per se , i . e . with positive expectations . Nevertheless , animal models , genetic studies and pharmacological manipulations suggest that serotonin plays a crucial role in the inhibition of active behaviours by aversive expectations [25] , [47] , [48] , [50] , [73] , [76]–[78] . In humans , there is evidence for the serotonergic mediation of the inhibition of active approach by aversive predictions [51] , and of approach responses to stimuli that are predictive of negative reinforcement [73] . It should be noted , though , that , acting via the indirect path and D2 receptors , dopamine itself has also been suggested to be important in mediating ‘nogo’ behaviour due to punishments [18] , [53] , [79] . Aversive Pavlovian stimuli can also potentiate behaviour [1] , [64] , [80] , [81] , with both serotonin and dopamine involved . Dopamine may have a dominant influence in this: it is both known to be released , and influential , in some aversive settings [82]–[85] and has a more evident relationship to vigour [33] , [34] . This observation has led to a re-interpretation of previous notions [43] of the opponency between dopamine and serotonin , putting an axis spanning invigoration and inhibition together with spanning reward and punishment [52] . Thus , the literature suggests three predictions for genetic correlates of the Pavlovian influences we observe . When considering these , the caveats concerning the interaction of genetic variation with psychopathology ( e . g . anxiety or depression ) , and with development need to be kept in mind . Nevertheless , the conditioned suppression effect of aversive Pavlovian stimuli on approach should be enhanced by D2 receptors , and hence be positively related to D2 striatal receptor density thought to be modulated by C975T ( rs6277; [17] ) . Second , conditioned suppression should be increased in subjects with higher serotonin levels , i . e . as might be the case with the less efficient ( s ) allelic variation of the serotonin reuptake transporter ( 5HTTLPR SLC6A4 [86] ) . Third , given dopamine's established positive correlation with approach and PIT [87] , [88] , we expect genetic polymorphisms that boost DA levels , such as the SLC6A3 polymorphism of the dopamine transporter [89] , to increase the impact of appetitive Pavlovian stimuli on approach . A similar effect may be expected from DARPP-32 , although its closer relationship to synaptic plasticity would also suggest effects on instrumental learning [90]–[92] . Although the learning parameters associated with instrumental approach and withdrawal did not differ , the impact of rewards and punishments on the acquisition of responding was highly asymmetric . In general , subjects neglected punishments , whilst maintaining a fixed sensitivity to reward . This was gratuitous as , in our setting , rewards and punishments were equally informative . It is , however , the case that the optimal strategy can be arrived at by concentrating on either . Subjects were not globally insensitive to punishments , as their choice behaviour in the Pavlovian learning was highly accurate both for rewards and punishments . Furthermore , it should be emphasized that ascribing punishments a value of zero outcome would still effectively behave as a punishment because a zero outcome is well below the average expectation of correct actions ( Figure 4D ) and as such would reduce the tendency to emit the action that caused it . The asymmetry has been noted before . Others have fitted models with separate learning rates for rewards and punishments and reported significantly slower learning rates for punishments than rewards [93] , [94] . In some restricted regimes , learning rates and inverse temperature parameters can trade off , and we explicitly tested both types of models to address this . One potential confound is the emergence of determinism . Subject were instructed to perform choices relative to mushrooms . Real world mushrooms are either edible or poisonous , and this dichotomy may have predisposed subjects towards a deterministic , rather than a matching , strategy . ( For instance , subjects may have chosen responses based on a classification of the mushrooms into ‘good’ and ‘bad’ ones , rather than on the particular value of a response for a mushroom . ) Indeed , in RL settings it is typically optimal to start with a low , exploratory , sensitivity to outcomes , but to increase this over time to encourage exploitation , culminating in a deterministic strategy [2] . However , subjects did not behave deterministically at any point ( Figure 2A ) and supplementary analyses showed that the time-varying pattern of reinforcement sensitivities this would predict is not observed in the data ( Text S1 ) . A further potential confound is the average stay probability . If this were precisely half-way between the stay probabilities after rewards and punishments in Figure 2B , then rewards and punishments would have the same effect relative to the baseline , and hence arguably be equally informative . However , this argument would neglect the fact that the mean stay probability itself must be a function of the reinforcement history; and that this must be included in making inferences about the reinforcement sensitivity . We have previously made the argument on theoretical grounds that part of the asymmetry observed in appetitive and aversive systems might be due to the inherent difference in how informative rewards and punishments are processed , enshrined again in the architecture of the striatum and neuromodulation [50] . Rewards tell us what to do; punishments tell us what not to do . The former is more informative in naturalistic settings where many options are available but only few are good . The fact that subjects gratuitously rely on rewards rather than on punishments in the present setting may reflect an implicit appreciation of this fact , although our findings are certainly in no way conclusive evidence . Interestingly , it is known that stronger optimality results can be shown for a stochastic learning automata rule called linear reward-inaction , which does not change propensities in the light of punishments but only rewards ( [95] , [96]; also known as a benevolent automaton [97] ) , than for a rule that changes propensities for both . The computational model served several central roles . First , it encapsulated the manifold aspects of behaviour and learning jointly , thereby controlling for them: the bias against withdrawals is not a due to a difference in learning; and variations in learning or generalization do not account for the PIT effects we saw . Secondly , its close fit to the behaviour argues that the PIT effects can be accounted for by a simple superposition of an instrumental and a Pavlovian controller: the action propensities due to both controllers were simply multiplied ( as additive factors in an exponential ) , rather than being allowed to interact in more complex ways . The simplicity of this interaction eschews questions about peripheral versus central response competition , whether appetitive and aversive systems compete centrally [7] , and whether Pavlovian learning is involved in instrumental learning [1] . It takes the view of multiple , separate controllers contributing in parallel [98] , and weighting the ultimate choice by the reward expected from that choice . One alternative would be to weigh contributions by different controllers according to their certainty [99] , although it is unclear how to compute the Pavlovian controller's certainty . There are various pressing directions for future studies . First , despite the role the architecture of decision-making has played in the argument , our work does not directly address the neural mechanisms concerned . These could be examined using imaging and pharmacological manipulations . Second , our task was not designed to distinguish between outcome-specific and general mechanisms [63] , [75] as we relied on one , monetary , outcome throughout . Studying different outcomes is important , given evidence for partly parallel pathways through different nuclei of the amygdala and different targets in the nucleus accumbens [100] , [101] . Third , we are missing one crucial further orthogonalization to do with the overall framing of the instrumental task . It is important to consider the case in which subjects can at best avoid losing money by doing the correct action [51] . We would expect punishment to maintain its instrumental force in this case; but there could also be a systematic difference in the nature of the Pavlovian influences . Pavlovian responses are believed to be hard-wired to reflect evolutionarily appropriate attitudes to predictions , being highly adaptive and sensitive to environmental structures [102] . Here , we showed that Pavlovian influences on instrumental behaviour depend on the intrinsic affective label of an action , independent of its learned reward expectation . It has long been known that prepared or compatible [27] , [69] behaviours are easier targets for instrumental conditioning . These intrinsic biases , or priors , may serve a crucial function both by reducing the need for collecting data ( i . e . sample complexity ) about the effects of actions , and by reducing the need for executing complex processing necessary to work out optimal actions ( i . e . computational complexity ) . Both of these can be expensive or dangerous , particularly in an aversive context . Our findings sharpen the understanding of the relative contribution of Pavlovian and instrumental contingencies in general tasks . We showed clearly that the interaction of Pavlovian and instrumental behaviours is organized along the lines of appetitive and aversive motivational systems , and that a critical contributor to this is the affective nature of actions .
54 healthy subjects of central European origin were recruited from the Berlin area . Subjects were screened for a personal history of neurological , endocrine , cardiac and psychiatric disorders ( SCID-I screening questionnaire ) , and for use of drugs and psychotropic medication in the past 6 months . Subjects received performance-dependent compensation ( 5–32 Euro ) for participation . Three subjects did not meet inclusion criteria and one subject did not complete the task; the data for three further subjects were lost due to a programming error . One further subject was excluded from the analysis because the instrumental task was not satisfactorily performed . The 46 remaining subjects were years old . 59% were female ( ) . The study was approved by the local Ethics Committee and was in accord with the Declaration of Helsinki 2008 . Subjects were given detailed information and gave written consent . They were seated comfortably at a table in front of a laptop with headphones and used a mouse with their dominant hand to indicate their choices . The amount earned was indicated by the computer , and the sum paid in cash at the end of the session . The computer task was followed by completion of self-rating scales . The task was written using Matlab and Psychtoolbox ( http://psychtoolbox . org ) . It consisted of one approach and one withdrawal block separated by a 2 minute break . Each block was in turn divided into a instrumental training , a Pavlovian training and a PIT part . Table 1 illustrates this . We modified a standard reinforcement learning model to capture the behavioural choices in the experiment . We first describe the main model , and then the alternative control models . Considering first the instrumental part , let be the instrumental stimulus ( out of up to 12; i . e . the subscript now designates time rather than identity as in Table 1 ) presented at trial , and the action ( choice ) on that trial . An action can be one of four types: go withdrawal and nogo withdrawal in the withdrawal block , and go approach and nogo approach in the approach block . Let also be the reinforcement obtained , either for a punishment , or for a reward . We write the probability of action in the presence of stimulus as a standard probabilistic function of i ) the reinforcement expectations associated with that pair on that trial , and ii ) a time-invariant , fixed , response bias : ( 1 ) ( 2 ) where is the instrumental weight of action , and where the variable can take on value for withdrawal go actions , or for the approach go actions . It is always zero for the nogo action . There was no delayed outcome in the instrumental task , and the expectations were thus constructed by a Rescorla-Wagner-like rule with a fixed learning rate . The immediate , intrinsic , value of the reinforcements delivered in the experiment may have different meaning for different subjects . To measure this effect , we added two further parameters: the reward sensitivity and the punishment sensitivity , yielding an update equation for the expectations:This is model 5 in Table 2 , which has the lowest score ( see below ) . Alternative models tested on the instrumental data only are as follows: Model 1 assumes that , and that . Model 2 allows only for separate reward and punishment sensitivities and model 4 for separate biases . Model 3 again assumes , and that , but allows for two separate learning rates , i . e . in Equation 3 is replaced by on trials where , and by on trials where . Model 6 and 7 are expansions of the final model , allowing for separate reward and punishment sensitivities ( model 6 ) and for separate learning rates ( model 7 ) in the approach and withdrawal conditions . Our main measure of interest is the effect of Pavlovian stimuli on the approach and withdrawal actions . Let additionally be the Pavlovian stimulus on trial . We can then write an equation similar to equation 2 for the trials where both instrumental and Pavlovian stimuli were present , but including a term that quantifies the effect of the particular Pavlovian stimulus on the action . This means that the action weights due to the instrumental and Pavlovian controllers are added inside the exponent of equation 2 , and that thus the probabilities each controller attaches to a particular action are multiplied and renormalized . The two controllers are therefore treated as two distinct entities , each separately voting for a particular action to be emitted . The influence of each system on action choice is relative to the strength with which the other enhances one particular action . We write the PIT weight of action as: ( 3 ) Here we force at all times . The go values can take on 10 separate , inferred , values , meaning that there is one separate parameter for each of the five Pavlovian stimuli in each of the two blocks . Each of these parameters captures how much boosts the go over the nogo action ( if ) or the inverse ( if ) . Note that because these are separately inferred , independent , parameters , this formulation does not impose any assumptions about the effect of the value of the stimulus , or about the relative effect of different stimuli with different values . Hence , this controls for variation in learning during the Pavlovian training block ( though the query trials indicate that learning was very robust ) . Equation 3 ( Model 8 in Table 2 ) assumes that the stimulus-action values at the end of the instrumental block are perfectly and exactly generalized to the PIT block . We first tested an alternative model ( Model 9 in Table 2 ) that included an exponential extinction factor , letting the values decay on each PIT trial by with . Next , we tested the model described in the main text ( Model 10 in Table 2 ) , which allowed for a fixed , Gaussian random offset between the effective values in the instrumental and PIT stages , i . e . we wrote:The noise factor took on value for the nogo action ( akin to the bias and variables ) . It took on a separate value—which was inferred as a separate parameter—for each subject and each stimulus . However , all stimuli shared the same prior distribution for this noise variable . That is , in the E step of our EM procedure , we fitted one Gaussian mean and variance to all the 's that had been inferred for all stimuli for all subjects . In this sense , the generalization factors were drawn from one Gaussian prior whose mean and variance were fitted just like the mean and variance of the other parameters . For each subject , each model specifies a vector of parameters . Assuming Gaussian prior distributions , we find the maximum a posteriori estimate of the parameters for each subject :where are all actions by the subject . We assume that actions are independent ( given the stimuli , which we omit for notational clarity ) , and thus factorize over trials . The prior distribution on the parameters mainly serves to regularise the inference and prevent parameters that are not well-constrained from taking on extreme values . We set the parameters of the prior distribution to the maximum likelihood given all the data by all the subjects:where . This maximisation is straightforwardly achieved by Expectation-Maximisation [109] . We use a Laplacian approximation for the E-step at the iteration:where denotes a normal distribution over with mean and is the second moment around , which approximates the variance , and thus the inverse of the certainty with which the parameter can be estimated . Finally , the hyperparameters are estimated by setting the mean and the ( factorized ) variance of the prior distribution to:ansformed before inference to enforce constraints . Unconstrained parameters are inferred in their native space . These model fitting procedures were verified on surrogate data generated from a known decision process . We fitted a large number of different models to the data , and some of these models differ in their flexibility . For instance , Model 8 , which assumes that the instrumental values are generalized exactly to the PIT stage is much less flexible than models 9–10 , which allow for an offset . It is important to choose that model which is flexible enough to explain the data , but not so flexible that it would also fit very different data equally well [109] . Ideally , this is achieved by computing the posterior log likelihood of each model given all the data . As we have no prior on the models themselves ( testing only models we believe are equally likely a priori ) , we instead examine the model log likelihood directly . This quantity can be approximated in two steps . First , the integral over [110]:Importantly , however , is not the sum of individual likelihoods , but in turn an integral over the parameters of each individual subject:The second line shows that we approximated the integrals by ( importance ) sampling times from the empirical prior distribution [109] . These samples were then also used to derive the error bars as the second moments around the maximum:where is a vector of zeros of the same dimension as with only entry set to one . The shifted likelihoods can be easily computed by re-weighting the samples drawn before:Note that while this model comparison procedure does give a good comparative measure of model fit , we still need an absolute measure to ensure that the best model does indeed provide a model fit that is adequate ( even the best might be bad ) . Given each subject's MAP parameter estimate , we compute the total “predictive probability”: ( 4 ) where we suppressed the dependence on stimuli on the LHS for clarity . We note that depends on the parameters , which have been fitted to the data . We term it a predictive probability in the sense that it predicts a subject's choice at time given that subject's past behaviour . We emphasize however , that this does depend on the MAP parameters fitted to that subjects' entire choice dataset . Finally , we test whether the expected number of choices predicted correctly exceeds that expected by chance ( using a binomial test ) . The overall predictive probability is given by the geometric mean over all choices and subjects: . | Beautiful background music in a shop may well tempt us to buy something we neither need nor want . Valenced stimuli have broad and profound influences on ongoing choice behaviour . After replicating known findings whereby approach is enhanced by appetitive Pavlovian stimuli and inhibited by aversive ones , we extend this to withdrawal behaviours , but critically controlling for the valence of the withdrawal behaviours themselves . We find that even when withdrawal is appetitively motivated , it is still inhibited by appetitive Pavlovian stimuli and enhanced by aversive ones . This shows , for the first time , that the effect of background Pavlovian stimuli depends critically on the intrinsic valence of behaviours , and differs between approach and withdrawal . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] | [
"neurobiology",
"of",
"disease",
"and",
"regeneration",
"behavioral",
"neuroscience",
"cognitive",
"neuroscience",
"computational",
"neuroscience",
"biology",
"neuroscience",
"animal",
"cognition"
] | 2011 | Disentangling the Roles of Approach, Activation and Valence in Instrumental and Pavlovian Responding |
Humans stand out from other animals in that they are able to explicitly report on the reliability of their internal operations . This ability , which is known as metacognition , is typically studied by asking people to report their confidence in the correctness of some decision . However , the computations underlying confidence reports remain unclear . In this paper , we present a fully Bayesian method for directly comparing models of confidence . Using a visual two-interval forced-choice task , we tested whether confidence reports reflect heuristic computations ( e . g . the magnitude of sensory data ) or Bayes optimal ones ( i . e . how likely a decision is to be correct given the sensory data ) . In a standard design in which subjects were first asked to make a decision , and only then gave their confidence , subjects were mostly Bayes optimal . In contrast , in a less-commonly used design in which subjects indicated their confidence and decision simultaneously , they were roughly equally likely to use the Bayes optimal strategy or to use a heuristic but suboptimal strategy . Our results suggest that , while people’s confidence reports can reflect Bayes optimal computations , even a small unusual twist or additional element of complexity can prevent optimality .
Humans and other animals use estimates about the reliability of their sensory data to guide behaviour ( e . g . [1–3] ) . For instance , a monkey will wait until its sensory data is deemed sufficiently reliable before taking a risky decision [3] . Humans can go further than other animals: they can explicitly communicate estimates of the reliability of their sensory data , by saying , for instance , “I’m sure”—an ability that is important for effective cooperation [4–6] . This ability to report on the reliability of our internal operations is known as “metacognition” , and is typically studied by asking people to report their confidence in the correctness of some decision [7] . However , the computations underlying confidence reports remain a matter of debate ( see Box 1 in [6] , for a brief overview ) . For instance , in an orientation-discrimination task , reports might—as a heuristic—reflect the perceived tilt of a bar . Alternatively , reports might reflect more sophisticated computations , like Bayesian inference about the probability that a decision is correct . An accurate understanding of confidence reports is important given their role in high-risk domains , such as financial investment ( e . g . [8] ) , medical diagnosis ( e . g . [9] ) , jury verdicts ( e . g . [10] ) , and politics ( e . g . [11] ) . Here , we ask: how do people compute their confidence in a decision ? We are particularly interested in whether confidence reports reflect heuristic or Bayes optimal computations . The latter would be consistent with a wide array of work showing that other aspects of perception and decision making are Bayes optimal [12] . However , as far as we know , whether confidence reports reflect Bayes optimal computations has not been directly tested . We use a standard psychophysical task in which subjects receive sensory data , make a decision based on this data , and report how confident they are that their decision is correct . Our goal is to determine how subjects transform sensory data into a confidence report . In essence , we are asking: if we use x to denote the sensory data ( x can be multi-dimensional ) and c to denote a confidence report , what is the mapping from x to c ? Alternatively , what is the function c ( x ) ? To answer this question , we follow an approach inspired by signal detection theory [13] . We hypothesize that subjects compute a continuous decision variable , zD ( x ) , and compare this variable to a single threshold to generate a decision , d . Likewise , we hypothesize that subjects compute a continuous confidence variable , zC ( x; d ) , an internal representation of the evidence in favour of the chosen decision , d , and compare this variable to a set of thresholds to generate a level of confidence , c ( the evidence in favour of one decision is different from the evidence in favour of the other decision , so the confidence variable must not only depend on the sensory evidence , x , but also the decision , d ) . Within this framework , a heuristic computation is a reasonable , but ultimately somewhat arbitrary , function of the sensory data . For instance , if the task is to choose the larger of two signals , x1 or x2 , a heuristic confidence variable might be the difference between the two signals: z Δ C C ( x ; d = 2 ) = x 2 − x 1 ( the subscript Δ denotes difference ) . The Bayes optimal confidence variable , on the other hand , is the probability that a correct decision has been made: z B C ( x ; d ) = P ( correct | x , d ) ( the subscript B denotes Bayesian ) . The question of whether confidence reports reflect Bayes optimal ( or simply Bayesian ) computations has important implications for inter-personal communication . In particular , probabilities , as generated by Bayes optimal computations , can easily be compared across different tasks ( e . g . perception versus general knowledge ) , making them easier to map onto reports . In contrast , heuristic computations typically lead to task-dependent internal representations , with ranges and distributions that depend strongly on the task , making it difficult to map them onto reports consistently , or compare them between different people . To our knowledge , it is impossible to determine directly the confidence variable , zC ( x; d ) ; instead , we can consider several models , and ask which is most consistent with experimental data . Choosing among different models for the confidence variable , zC ( x; d ) , is straightforward in principle , but there are some subtleties . The most important subtlety is that if the task is “too simple” , it is impossible to distinguish one model from another . Here , “too simple” means that the sensory data , x , consists of a single signal , which we write x to indicate that it is scalar . To see why , let’s say we wanted to distinguish between some heuristic confidence variable , say z H C ( x ; d ) = x , and the Bayes optimal confidence variable , z B C ( x ; d ) = P ( correct | x , d ) . Suppose we found empirically that a subject reported low confidence when the heuristic variable , z H C ( x ; d ) , was less than 0 . 3 and high confidence when the heuristic variable was greater than 0 . 3 . Clearly there is a deterministic mapping from the heuristic variable to the confidence reports , but is it in any way unique ? The answer is no . For example , if the Bayesian variable is greater than 0 . 4 whenever the heuristic variable is greater than 0 . 3 , then it is also true that our subject reported low confidence when the Bayesian variable was less than 0 . 4 and high confidence when the Bayesian variable was greater than 0 . 4 . Thus , there is absolutely no way of knowing whether our subjects’ confidence reports reflect the heuristic or the Bayesian confidence variable . In general , there is no way to distinguish between any two functions of x that are monotonically related—one can simply map the thresholds through the relevant function , as shown in Fig 1 . The situation is very different when x is a vector ( i . e . two or more sensory signals ) . As in the one-dimensional case , consider two models: a heuristic model , z H C ( x ; d ) , and a Bayes optimal model , z B C ( x ; d ) . In general , if x is a vector , it is not possible to get the same mapping from x to c using z H C ( x ; d ) and z B C ( x ; d ) . In particular , when z H C ( x ; d ) and z B C ( x ; d ) provide a different ordering of the x’s—whenever we have z H C ( x 1 ; d ) > z H C ( x 2 ; d ) and simultaneously z B C ( x 1 ; d ) < z B C ( x 2 ; d ) —then it is not possible to find pairs of thresholds that lead to the same region in x-space . Thus , although we cannot say much about the confidence variable for one-dimensional signals , we can draw strong conclusions for multi-dimensional signals . This difference between one-dimensional and multi-dimensional sensory data is one of the key differences between our work and most prior work . Previous models based on signal detection theory have typically assumed that the sensory data is one-dimensional ( e . g . [14–16] ) , leaving them susceptible to the problem described above . There is also a variety of “dynamic” signal detection theory models in which sensory data is assumed to accumulate over time ( see Pleskac & Busemeyer ( 2010 ) [17] , for an overview ) . Such models are able to explain the interplay between accuracy , confidence , and reaction time—something that we leave for future work . However , in these models , the sensory data is also summarised by a single scalar value , making it impossible to determine whether subjects’ confidence reports reflect heuristic or Bayes optimal computations . Here we considered multi-dimensional stimuli in a way that allows us to directly test whether subjects’ confidence reports reflect heuristic or Bayes optimal computations . In our study , subjects were asked to report their confidence in a visual two-interval forced-choice task . This allowed us to model the sensory data as having two dimensions , with one dimension coming from the first interval and the other from the second interval . We considered three models for how subjects generated their confidence—all three models were different “static” versions of the popular race model in which confidence reports are assumed to reflect the balance of evidence between two competing accumulators ( originally proposed by Vickers ( 1979 ) [18] , and more recently used in studies such as Kepecs et al . ( 2008 ) [1] , and de Martino et al . ( 2013 ) [19] ) . The first model , the Difference model , assumed—in line with previous work—that subjects’ confidence reports reflected the difference in magnitude between the sensory data from each interval . The second model , the Max model , assumed that subjects’ confidence reports reflected only the magnitude of the sensory data from the interval selected on a given trial—thus implementing a “winner-take-all” dynamic [20] . The third model , the Bayes optimal model , assumed that subjects’ confidence reports reflected the probability that their decision was correct given the sensory data from each interval . Furthermore , we tested two different methods for eliciting confidence—both being used in research on metacognition [7] . In the standard two-response design , subjects first reported their decision , and only then , and on a separate scale , reported their confidence . In the less-commonly used one-response design , subjects reported their confidence and decision simultaneously on a single scale . We were interested to see whether the more complex one-response design—in which subjects , in effect , have to perform two tasks at the same time—affected the computations underlying confidence reports as expected under theories of cognitive load ( e . g . [21 , 22] ) and dual-task interference ( e . g . [23 , 24] ) . We used Bayesian model selection to assess how well the models fit our data; thus our analysis was “doubly Bayesian” in that we used Bayesian model selection to test whether our subjects’ behaviour was best explained by a Bayes optimal model [25] . We found that the commonly used Difference model was the least probable model irrespective of task design . Subjects’ confidence reports in the two-response design were far more likely to reflect the Bayes optimal model rather than either heuristic model . In contrast , in the one-response design , the confidence reports of roughly half of the subjects were in line with the Bayes optimal model , and the confidence reports of the other half were in line with the Max model , indicating that , perhaps , the increased cognitive load in the one-response paradigm caused subjects to behave suboptimally . In sum , our results indicate that while it is possible to generate confidence reports using Bayes optimal computations , it is not automatic—and can be promoted by certain types of task .
Participants were undergraduate and graduate students at the University of Oxford . 26 participants aged 18–30 took part in the study . All participants had normal or corrected-to-normal vision . The local ethics committee approved the study , and all participants provided written informed consent . To model responses , we assumed the following: On each trial , subjects receive a pair of sensory signals , x . Subjects transform those sensory signals into a continuous decision variable , zD ( x ) , and then compare this variable to a single threshold to make a decision , d . Finally , subjects transform the sensory signals and the decision into a continuous confidence variable , zC ( x; d ) , and then compare this variable to a set of thresholds to obtain a confidence report , c . This section starts by describing our assumptions about the sensory signals , x , then moves on to the models for how subjects might compute their decision and confidence variables . Finally , we describe the Bayesian inference technique used to fit the parameters and find the most probable model . We wish to compute the probability of the various models given our data . The required probability is , via Bayes’ theorem , P ( m|data ) ∝P ( m ) P ( data|m ) ( 12 ) where m is either Δ ( Difference model ) , M ( Max model ) or B ( Bayesian model ) . The data from subject l consists of two experimenter-defined variables: the target intervals , il , and the target contrasts , sl , and two subject-defined variables: the subject’s decisions , dl , and the subject’s confidence reports , cl . Here , the bold symbols denote a vector , listing the value of that variable on every trial; for instance the interval on the kth trial is ilk . We fit different parameters to every subject , so the full likelihood , P ( data∣m ) , is given by a product of single-subject likelihoods , P ( data|m ) =∏lP ( dl , cl , il , sl|m ) . ( 13 ) Because il and sl are independent of the model , m , we may write P ( data|m ) ∝∏lP ( dl , cl|il , sl , m ) . ( 14 ) To compute the single-subject likelihood we cannot simply choose one setting for the parameters , because the data does not pin down the exact value of the parameters . Instead we integrate over possible parameter settings , P ( dl , cl|il , sl , m ) =∫P ( dl , cl|il , sl , m , θl , σl , bl ) P ( θl ) P ( σl ) P ( bl ) dθldσldbl , ( 15 ) where θl collects that subject’s decision and confidence thresholds . This integral is large if the best fitting parameters explain the data well ( i . e . if P ( dl , cl∣il , sl , m , θl , σl , bl ) is large for the best fitting parameters ) , as one might expect . However , this integral also takes into account a second important factor , the robustness of the model . In particular , a good model is not overly sensitive to the exact settings of the parameters—so you can perturb the parameters away from the best values , and still fit the data reasonably well . This integral optimally combines these two contributions: how well the best fitting model explains the data , and the model’s robustness . For a single subject ( dropping the subject index , l , for simplicity , but still fitting different parameters for each subject ) , the probability of d and c given that subject’s parameters is the product of terms from each trial , P ( d , c | i , s , m , θ , σ , b ) = ∏ k P ( d k , c k | i k , s k , m , θ , σ , b ) , ( 16 ) We therefore need to compute the probability of a subject making a decision , dk , and choosing a confidence level , ck , given the subject’s parameters , the target interval , ik , and target contrast , sk . We do this numerically , by sampling: given a set of parameters , θ , σ and b we generate an x from either Eqs ( 1 ) or ( 2 ) ( depending on whether ik is 1 or 2 ) . We compute zD ( x ) from either Eqs ( 3 ) , ( 5 ) or ( 7 ) ( depending on the model ) , and threshold zD ( x ) to get a decision , d . Next , we combine x and d to compute zC ( x; d ) from either Eqs ( 4 ) , ( 6 ) or ( 8 ) ( again , depending on the model ) , and threshold zC ( x; d ) to get a confidence report , c . We do this many times ( 105 in our simulations ) ; P ( dk , ck∣ik , sk , m , θ , σ , b ) is proportion of times the above procedure yields d = dk and c = ck . To perform the integral in Eq ( 15 ) , we must specify prior distributions over the parameters σ , b and θ . While it is straightforward to write down sensible priors over two of these parameters , σ and b , it is much more difficult to write down a sensible prior for the thresholds , θ . This difficulty arises because the thresholds depend on zD ( x ) and zC ( x; d ) , which change drastically from model to model . To get around this difficulty , we reparametrise the thresholds , as described in the next section .
To compare models , we look at the posterior probability of each of our models given the data , P ( m∣data ) . As , a-priori , we have no reason to prefer one model over another , we use a uniform prior , P ( m ) = 1/3 , so , assuming that every subject uses the same model , then the posterior is proportional to P ( data∣m ) , which we showed how to compute in the Model Comparison Section . The Bayesian model is better by a factor of around 104 for the one-response data and around 1025 for the two-response data ( Fig 4 ) . For the above model comparison , we assumed that all subjects used the same model to generate their confidence reports . It is quite possible , however , that different subjects use different models to generate their confidence reports . In particular , we might expect that there is some probability with which a random subject uses each model , P ( ml ) ( where l is the subject index , so ml is the model chosen by subject l ) . Under this assumption , we can analyse how well the models fit the data by inferring the probability with which subjects choose to use each model , P ( ml ) , using a variational Bayesian method presented by [26] . In agreement with the previous analysis , we find that for the two-response dataset , the probability of any subject using the Bayesian model is high: subjects are significantly more likely to use the Bayesian model than either the Max or Difference models ( p < 0 . 006; exceedence probability [26]; Fig 5B ) . For the one-response dataset , on the other hand , subjects use the Bayesian model only slightly more than the Max model ( Fig 5A ) . The log-likelihood differences for individual subjects are plotted in Fig 5C and 5D , with uncertainty given by the size of the crosses . Again , for the two-response dataset , but not for the one-response dataset , the difference between each subject’s log-likelihood for the Bayesian and Max models is larger than 0 ( two-response: t ( 10 ) = 3 . 47 , p < . 006; one-response: t ( 14 ) = 0 . 954 , p ≈ . 35; two-sided one-sample t-test ) . While the model evidence is the right way to compare models , it is important to check that the inferred models and parameter settings ( for inferred parameters for each subject see S1 and S2 Tables ) are plausible . We therefore plotted the raw data—the number of times a participant reported a particular decision and confidence level for a particular target interval and target contrast—along with the predictions from the Bayesian model . In particular , in Fig 6 , we plot fitted and empirical distributions over confidence reports given a target interval and contrast from an example participant ( for all subjects and all models see S1 and S2 Figs ) . To make this comparison , we defined “signed confidence” , whose absolute value gives the confidence level , and whose sign gives the decision , Signed confidence = { - c for d = 1 - c for d = 2 . ( 29 ) These plots show that our model is , at least , plausible , and highlights the fact that our model selection procedure is able to find extremely subtle differences between models . Plotting psychometric curves ( Fig 7 ) gave similar results . Again , to plot psychometric curves , we defined “signed contrast” , whose absolute value gives the contrast , and whose sign gives the target interval , Signed contrast = { - s for i = 1 - s for i = 2 . ( 30 ) For model selection to actually work , there need to be differences between the predictions made by the three models . Here , we show that the models do indeed make different predictions under representative settings for the parameters . To understand which predictions are most relevant , we have to think about exactly what form our data takes . In our experiment , we present subjects with a target in one of the two intervals , i , with one of four contrast levels , s , then observe their decision , d and confidence report , c . Overall , we therefore obtain an empirical estimate of each subject’s distribution over decision and confidence reports ( or equivalently signed confidence , see previous section ) , given a target interval and contrast . This suggests that we should examine the predictions that each model makes about each subject’s distribution over decisions and confidence reports , given the target interval , i , and contrast , s . While these distributions are superficially very similar ( Fig 8 ) , closer examination reveals two interesting , albeit small , differences . Importantly , these plots display theoretical , and hence noise-free results , so even small differences are meaningful , and are not fluctuations due to noise . First , the Max model differs from the other two models at intermediate contrast levels , especially s = 0 . 07 , where the Max model displays bimodality in the confidence distribution . In particular , and unexpectedly , an error with confidence level 1 is less likely than an error with confidence levels 2 to 4 . In contrast , the other models display smooth , unimodal behaviour across the different confidence levels . This pattern arises because the Max model uses only one of the two sensory signals . For example , when s = 0 . 07 and i = 2 ( so the target is fairly easy to see , and is in interval 2 ) , then x2 is usually large . Therefore , for x1 to be larger than x2 , prompting an error , x1 must also be large . Under the Max model , x1 being large implies high confidence , and , in this case , a high confidence error . Second , the three models exist on a continuum , with the Max model using the narrowest range of confidence levels , the Bayesian model using an intermediate range , and the Difference model using the broadest range . These trends are particularly evident at the lowest and highest contrast levels . At the lowest contrast level , s = 0 . 015 , the distribution for the Max model is more peaked , whereas the distribution for the Difference model is lower and broader , and the Bayesian model lies somewhere between them . At the highest contrast level , s = 0 . 15 , the Max model decays most rapidly , followed by the Bayesian model , and then the Difference model . To understand this apparent continuum , we need to look at how the models map sensory data , defined by x1 and x2 , onto confidence reports . We therefore plotted black contours dividing the regions of sensory-space ( i . e . ( x1 , x2 ) -space ) that map to different confidence levels ( Fig 9 ) . These plots highlight striking differences between the models . In particular , the Difference model has diagonal contours , whereas the Max model has contours that run horizontally , vertically or along the central diagonal at x1 = x2 . In further contrast , the Bayesian model has curved contours with a shape somewhere between the Difference and Max models . In particular , for large values of x1 and x2 , the contours are almost diagonal , as in the Difference model whereas for small values of x1 and x2 , the contours are more horizontally or vertically aligned , as in the Max model . To see how differences in the mapping from sensory-space to confidence reports translate into differences in the probability distribution over confidence reports , we consider the red dots , representing different target intervals and contrasts . For instance , a high-contrast target in interval 2 ( s = 0 . 15 ) , is represented by the uppermost red dot in each subplot . Importantly , red dots representing stimuli lie along the horizontal and vertical axes ( green ) . The angle at which the contours cross these axes therefore becomes critically important . In particular , for the Difference model the contours pass diagonally through the axes , and therefore close to many red dots ( representing stimuli ) , giving a relatively broad range of confidence levels for each stimulus type . In contrast , for the Max model , the contours pass perpendicularly through the axes , minimizing the number of red dots ( representing stimuli ) that each contour passes close to , giving a narrower range of confidence levels for each stimulus type . The contours of the Bayesian model pass through the axes at an angle between the extremes of the Difference and Max models—as expected , giving rise to a range of confidence levels between the extremes of the Difference and the Max model . In principle , these differences might allow us to choose between models based only on visual inspection of P ( d , c∣i , s , m , params ) . However , in practice , the distribution over decision and confidence reports , averaging over trial type , pd , c , is not constant , as we assumed above , but is far more complicated . This additional complexity makes it impossible to find the correct model by simple visual inspection . More powerful methods , like Bayesian model selection , are needed to pick out these differences .
Barthelmé & Mamassian ( 2009 ) [27] went part-way towards realizing the potential of using multidimensional stimuli . Subjects were asked to indicate which of two Gabor patches they would prefer to make an orientation judgement about . Interestingly , and in contrast to our results , they found that subjects were more likely to use a heuristic strategy ( similar to the Max model ) than a Bayes optimal strategy . However , there were three aspects of their study that make it potentially less relevant to the question of whether confidence reports reflects Bayes optimal computations . First , our model selection procedure is fully Bayesian , and therefore takes account of uncertainty in model predictions , whereas their procedure was not . In particular , under some circumstances a model will make strong predictions ( e . g . “the subject must make this decision” ) , whereas under other circumstances , the model might make weaker predictions ( e . g . “the subject is most likely to make this decision , but I’m not sure—they could also do other things” ) . Bayesian model selection takes into account the strength or weakness of a prediction . Second , in real life ( and in our study ) , people tend to report confidence using verbal ( e . g . “not sure” to “very” sure ) or numerical ( e . g . 1 to 10 ) scales . In contrast , in Barthelmé & Mamassian ( 2009 ) [27] , subjects simply made a forced choice between two stimuli . Third , in their study , the Difference model made exactly the same predictions as the Bayes optimal model , making it impossible to distinguish these computations . There are , of course , other approaches for addressing the question of whether the confidence variable is Bayes optimal . Barthelmé & Mamassian ( 2010 ) [28] showed that subject’s confidence variable can take into account two factors ( contrast and crowding ) that might lead to uncertainty—as opposed to using only one factor . Similarly , de Gardelle & Mamassian ( 2014 ) [29] showed that subjects were able to accurately compare the confidence variable across different classes of stimuli ( in this case orientation discrimination versus spatial frequency discrimination ) . These studies provide some , albeit indirect , evidence that confidence reports might indeed reflect probability correct , in agreement with our work . Confidence reports have been observed to vary with a range of factors that we did not consider here . For example , people have been shown to be overconfident about the accuracy of their knowledge-based judgements , but underconfident about the accuracy of their perceptual judgements ( see [30] for a review ) . People’s general level of confidence may also vary with social context . When groups of people resolve disagreement , the opinions expressed with higher confidence tend to carry more weight ( e . g . [31] ) , so group members tend to increase their confidence to maximize their influence on the group decision [32 , 33] . They may also adjust their confidence reports to indicate submission or dominance , or cut their losses if they should turn out to be wrong ( e . g . [34] ) . Lastly , people’s confidence reports may vary with more general social factors such as profession , gender and culture: finance professionals are more confident than the average population ( e . g . [8] ) ; men are more confident than women ( e . g . [35] ) ; and people from Western cultures are more confident than people from East Asian cultures ( e . g . [36] ) . Our method allows us to think about the variability in confidence reports as having two dimensions . The first ( perhaps more superficial ) dimension relates to the average confidence level , or confidence distribution . We might imagine that this dimension is primarily modulated by social context , as described above . The second ( perhaps deeper ) dimension relates to the computations underlying confidence reports . In our data , there do indeed appear to be individual differences in how people generate their confidence reports , and very subtle changes to the task appear to affect this process . We might therefore expect shifts in how people generate their confidence reports for tasks of different complexity . For example , it is not straightforward to solve general-knowledge questions , such as “What is the capital of Brazil ? ” , using Bayesian inference . While one could in principle compute the probability that one’s answer is correct , the computational load may be so high that people resort to heuristic computations ( e . g . using the population size of the reported city ) . Future research should seek to identify how confidence reports change between task domains and social contexts—in particular , whether such changes are mostly due to changes in the computation used to generate the confidence variable ( cf . , zC ( x; d ) ) , or due to changes in the mapping of this variable onto some confidence scale . Many studies have asked whether confidence reports , and hence metacognitive ability , are optimal ( see [37] , for a review of measures of metacognitive ability ) . However , our work suggests that there are ( at least ) two kinds of optimality . First , the transformation of incoming data into an internal confidence variable ( i . e . zC ( x; d ) ) could be optimal—that is , computed using Bayesian inference . Second , the mapping of the confidence variable onto some external scale of confidence could be optimal ( i . e . c ( zC ( x; d ) ) ) , but this depends entirely on the details of the task at hand . For instance , without some incentive structure , there is no reason why subjects should opt for any particular mapping , as long as their mapping is monotonic ( i . e . reported confidence increases strictly with their confidence variable ) . Importantly , it does not seem that subjects use an optimal mapping , as evidenced by the large amount of research on “poor calibration”—that is , the extent to which the reported probability of being correct matches the objective probability of being correct for a given decision problem ( e . g . [30 , 38] ) . Even when there is an incentive structure , subjects only improve their calibration and never reach perfection ( e . g . [34 , 39] ) . Future research should seek to identify why poor calibration arises , and how it can be corrected . We asked how people generate their confidence reports . Do they take a heuristic approach , and compute some reasonable , but ultimately arbitrary , function of the sensory input , or do they take a more principled approach , and compute the probability that they are correct using Bayesian inference ? When subjects first made a decision and then reported their confidence in that decision , we found that their confidence reports overwhelmingly reflected the Bayesian strategy . However , when subjects simultaneously made a decision and reported confidence , we found the confidence reports of around half of the subjects were better explained by the Bayesian strategy , while the confidence reports of the other half of the subjects were better explained by a heuristic strategy . | Confidence plays a key role in group interactions: when people express an opinion , they almost always communicate—either implicitly or explicitly—their confidence , and the degree of confidence has a strong effect on listeners . Understanding both how confidence is generated and how it is interpreted are therefore critical for understanding group interactions . Here we ask: how do people generate their confidence ? A priori , they could use a heuristic strategy ( e . g . their confidence could scale more or less with the magnitude of the sensory data ) or what we take to be an optimal strategy ( i . e . their confidence is a function of the probability that their opinion is correct ) . We found , using Bayesian model selection , that confidence reports reflect probability correct , at least in more standard experimental designs . If this result extends to other domains , it would provide a relatively simple interpretation of confidence , and thus greatly extend our understanding of group interactions . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [] | 2015 | Doubly Bayesian Analysis of Confidence in Perceptual Decision-Making |
There are currently huge efforts by the World Health Organization and partners to complete global polio eradication . With the significant decline in poliomyelitis cases due to wild poliovirus in recent years , rare cases related to the use of live-attenuated oral polio vaccine assume greater importance . Poliovirus strains in the oral vaccine are known to quickly revert to neurovirulent phenotype following replication in humans after immunisation . These strains can transmit from person to person leading to poliomyelitis outbreaks and can replicate for long periods of time in immunodeficient individuals leading to paralysis or chronic infection , with currently no effective treatment to stop excretion from these patients . Here , we describe an individual who has been excreting type 2 vaccine-derived poliovirus for twenty eight years as estimated by the molecular clock established with VP1 capsid gene nucleotide sequences of serial isolates . This represents by far the longest period of excretion described from such a patient who is the only identified individual known to be excreting highly evolved vaccine-derived poliovirus at present . Using a range of in vivo and in vitro assays we show that the viruses are very virulent , antigenically drifted and excreted at high titre suggesting that such chronic excreters pose an obvious risk to the eradication programme . Our results in virus neutralization assays with human sera and immunisation-challenge experiments using transgenic mice expressing the human poliovirus receptor indicate that while maintaining high immunisation coverage will likely confer protection against paralytic disease caused by these viruses , significant changes in immunisation strategies might be required to effectively stop their occurrence and potential widespread transmission . Eventually , new stable live-attenuated polio vaccines with no risk of reversion might be required to respond to any poliovirus isolation in the post-eradication era .
Despite difficulties in interrupting wild poliovirus transmission in the last few remaining endemic countries and recent drawbacks due to international spread of poliovirus in central Asia , central Africa and the Middle East [1] , the global polio eradication appears to be within reach . Four of the six WHO regions have been certified polio-free and a country such as India , where massive poliomyelitis outbreaks were very common , interrupted circulation of endemic wild poliovirus in 2010 . There has been no case of poliomyelitis caused by circulating wild type 2 poliovirus since 1999 , no case of type 3 since November 2012 and the last case of type 1 in Africa was in August 2014 , leaving some areas of Pakistan and Afghanistan as the main remaining reservoirs [2] . All type 2 poliomyelitis cases since 1999 , except an isolated incident of 10 cases linked to a wild laboratory reference strain in India [3] , are due to vaccine-related poliovirus strains in either recipients , their immediate contacts or after the vaccine virus has regained the ability to transmit and circulate freely . Vaccine-associated paralytic poliomyelitis occurs in a very small proportion of vaccinees [4] and can be prevented by using inactivated rather than live vaccine . Vaccine-derived poliovirus ( VDPV ) strains , defined as those with more than 1% ( 0 . 6% for serotype 2 poliovirus ) sequence drift in the capsid VP1 gene with respect to the corresponding Sabin strain , can be generated and transmitted from person to person in populations with low immunity and have been associated with a number of poliomyelitis outbreaks around the world [5–9] . These circulating VDPVs ( cVDPVs ) behave very similarly to wild polioviruses and should therefore be eliminated by the same immunisation methods . In addition , some hypogammaglobulinaemic patients are known to excrete poliovirus for prolonged periods of time [10–12] but there is currently no effective strategy to deal with this problem . Although there has been some evidence of local virus transmission from these patients to unvaccinated children [13] , VDPV strains from immunodeficient individuals ( iVDPVs ) have not yet been implicated in outbreaks in the same way that cVDPVs have [14] . The World Health Organization ( WHO ) and partners have prepared endgame plans for the global polio eradication initiative ( GPEI ) which include the elimination of the serotype 2 component from the Sabin live-attenuated oral poliovaccine ( OPV ) and the implementation of global use of inactivated poliovaccine ( IPV ) [15] . This represents a major change after more than 50 years of trivalent OPV use for routine immunisation although monovalent and bivalent vaccines are commonly used for campaigns on national immunisation days . The risks posed by iVDPV strains and the prevalence of such cases globally are unknown so their relevance in the context of the GPEI endgame is not easy to assess . In order to better understand the growth and properties of iVDPV strains and their potential for transmission , we have characterised iVDPV isolates from an immunodeficient individual obtained during a period of more than 20 years . Although examples of long-term poliovirus excretion have been described before by us and others ( reviewed in [14] ) , they have mostly included a small number of samples from paralytic cases as otherwise asymptomatic long-term excreters remain undetected . In previous cases patients died , stopped shedding virus or were lost to follow up relatively soon after the first virus isolation . Important gaps in the scientific knowledge of long-term poliovirus excretion by these individuals remained such as determining changes in excretion titres , antigenic structure and neurovirulence of poliovirus following many years of evolution in a single individual as well as estimating the efficacy of current vaccines at preventing paralysis and transmission induced by these viruses . Our paper provides relevant findings in these areas that indicate that VDPV isolates form these patients represent a real risk of polio re-emergence in the post-eradication era , particularly considering there is currently no effective strategy to treat these patients .
The first stool samples from this individual were tested between March and November 1995 . At that time , type 2 VDPV isolates differing from the parental Sabin 2 OPV strain at between 9 . 9% and 11 . 3% of VP1 nucleotides were identified . A total of 185 subsequent samples have been obtained so far in the following years , all positive for iVDPV2 strains with virus titres shed in the stools typically around or above 4 log10 infectious particles per gram , comparable to virus titres shed by healthy vaccinees and paralytic cases infected with vaccine or wild poliovirus [16] . The latest isolate available was from 4th March 2015 showing a 17 . 7% VP1 sequence drift from Sabin 2 poliovirus . Phylogenetic analyses in the capsid region confirmed that the iVDPV strains were genetically related , sequentially evolved from Sabin 2 and distinct from other type 2 VDPVs and wild polioviruses ( Fig 1 ) . A Bayesian Monte Carlo Markov Chain ( MCMC ) phylogenetic analysis determined a mean evolutionary rate of 1 . 51×10−2 total substitutions/site/year [95% High Probability Distribution ( HPD95 ) range = 1 . 26–1 . 77×10−2] in the VP1 gene , similar to previous estimates for poliovirus VP1 [17] . The date of the initiating OPV dose was estimated to be 11th March 1986 [HPD95 = 6th July 1983-11th January 1989] , relatively close to 4th August 1986 , the date of the patient’s last known OPV vaccination . It is therefore most likely that this individual has been excreting poliovirus for around 28 years . There was no apparent effect on the virus evolution rate suggesting bottleneck effects due to the anti-viral treatments that failed to interrupt virus excretion from this patient [18] . However , a much more detailed analysis of virus population dynamics should be conducted to determine any possible effect due to the different anti-viral interventions . All iVDPV isolates showed reversion at the two known attenuation mutations of Sabin 2 vaccine strain: nucleotide 481 ( from A to G ) in the 5’ non-coding region ( 5’NCR ) and capsid amino acid VP1-143 ( from Isoleucine to Threonine ) and were highly neurovirulent in transgenic mice expressing the human poliovirus receptor . The 50% paralytic dose ( PD50 ) values were comparable to those determined for cVDPV and wild polioviruses while the Sabin 2 vaccine strain did not paralyse any animals at the highest dose that could be given ( Fig 2 ) . Amino acid differences with respect to the Sabin 2 parental strain in the complete coding sequence of selected iVDPV isolates were determined ( S1 Table ) . All iVDPV strains contained identical changes from Sabin 2 at 52 amino acid positions . Forty mutations were present in at least two iVDPV isolates and 24 amino acid changes were unique . The proportion of nucleotide mutations leading to amino acid changes was high for all iVDPV strains . This contrasts with the low proportion of non-synonymous changes from Sabin 2 identified in cVDPV strains and wild isolates as found here and elsewhere , particularly in capsid sequences . It is not clear whether any of the numerous additional mutations incorporated in the iVDPV isolates have any effect on neurovirulence but they do not seem to have an overall mitigation impact for any of the isolates tested as it has been reported for one highly drifted type 2 VDPV isolate found in a sewage sample in Israel [19] . Many of the sequence changes between the iVDPV strains and the Sabin 2 virus resulted in amino acid differences in known antigenic sites [20] ( Table 1 ) . As a consequence , the iVDPV strains did not react at all with monoclonal antibodies against most of the known neutralising antibody sites ( Fig 3 ) . It was of interest that all isolates tested did react with antibodies specific for antigenic site 3b ( 1102 and 1103 ) . In contrast , the wild polioviruses strains analysed , which span almost four decades in time and which were isolated in geographically distant locations , exhibited an antigenic structure much closer to that of Sabin 2 virus , reacting with at least one monoclonal antibody specific for each antigenic site ( Fig 3 ) . A cVDPV strain from Madagascar [21] also reacted with most monoclonal antibodies . There was no evidence of sequences derived from Sabin 1 or Sabin 3 poliovirus vaccine strains nor sequences derived from other polio or non-polio human enterovirus isolates in any iVDPV genome examined . There was therefore no indication of recombination with other enteroviruses , although recombination within the iVDPV population is quite possible [22 , 23] . In contrast , virtually all cVDPV and wild type polio strains are recombinants with other group C enteroviruses and include sequences from the 5’NCR and/or the non-structural coding region [21 , 24] . Despite the extensive antigenic changes found in iVDPV strains ( Fig 4 ) , human sera readily neutralised iVDPV isolate 160198 , the most antigenically divergent strain ( Fig 5 ) . This isolate was the only one obtained by plaque purification so it might represent a minor variant with slightly different antigenic makeup . The log2 geometric mean titre ( GMT ) of antibodies neutralizing iVDPV virus 160198 in 40 serum samples from UK adults was 9 . 96±2 . 78 comparable to that against MEF-1 ( log2 GMT = 10 . 20±2 . 40 ) , the wild poliovirus strain used for IPV production , and Sabin 2 ( log2 GMT = 10 . 49±2 . 02 ) , used for OPV production . These differences were not statistically significant ( P = 0 . 95 for iVDPV vs MEF-1 and P = 0 . 36 for iVDPV vs Sabin2 , for paired results using the Wilcoxon signed-rank test ) . The results suggest that antibodies to antigenic site 3b in human sera , partially conserved in iVDPV strains , may be sufficient to neutralise the virus . Alternatively , other conserved antigenic epitopes not detected by our murine antibody panel but present in the iVDPV strains , could have contributed to the high neutralization levels shown in human sera . These results are reassuring in that they indicate that vaccinated humans are well protected against infection with these highly drifted iVDPV strains . However , the sera tested here correspond to a selected group of UK healthy adults between 28–65 years of age who had been vaccinated with a full course of four OPV doses plus at least one dose of IPV . The UK switched from OPV to IPV for polio immunisation in 2004 so it would be helpful to test sera from cohorts that have only received IPV immunisation . Israel , which also switched from OPV to IPV at a similar date ( 2005 ) , has recently detected the widespread circulation of type 1 wild poliovirus through environmental surveillance . There were no paralytic cases but , like the iVDPVs reported here , isolates from Israel showed antigenic differences from the corresponding vaccine strain which may have contributed to their ability to circulate in the context of IPV immunity [26] . We used a transgenic mouse model [27] to test the ability of different IPV products to protect against paralysis caused by iVDPV strains . Both conventional IPV ( cIPV ) based on wild poliovirus strains and Sabin IPV ( sIPV ) based on OPV strains were used ( Fig 6 ) . All unimmunised control mice were severely paralysed by 5 days ( MEF-1 virus ) or 8 days ( iVDPV strain ) post-challenge . All three cIPV products protected against both challenge strains with only one animal developing paralysis . In contrast , sIPV products were less effective at protecting mice and showed variable responses between them . For all vaccines , the in vitro neutralizing antibody titers in sera from immunized mice were at least 7-fold lower against the iVDPV strain than they were against MEF-1 virus ( Table 2 ) . Our results highlight the need for improving the standardisation of sIPV products in terms of measuring vaccine potency and defining the protective human dose . The lower immunogenicity shown by type 2 sIPVs has been reported before [28 , 29] . In conclusion , we describe a patient who has been excreting highly virulent and antigenically modified type 2 poliovirus at high titres for a period estimated to be twenty eight years so far . This is by far the longest reported poliovirus excretion and represents the most comprehensive collection of iVDPV sequential isolates available . Provided antibody titres and immunisation coverage are maintained it is likely that the population will be protected against paralytic disease , but it is also possible that this virus could circulate in populations only using IPV as described in Israel for wild poliovirus [26] , thus representing a possible source of polio re-emergence , particularly as these iVDPV strains are antigenically atypical and drifted from both Sabin 2 and MEF-1 vaccine strains . This is particularly relevant at present as there are imminent plans to remove type 2 poliovirus from OPV [15] . Moreover the use of IPV based on the Sabin strains is being encouraged by WHO for reasons of environmental safety and the data presented here suggest that it is less effective . Of the total of 73 iVDPV cases that have been described between 1962 and 2014 [14] , only seven of them involved infections lasting more than five years . The case described here represents the only individual of those seven known to be excreting at present . However , several highly drifted VDPV strains have recently been isolated from sewage samples in Slovakia , Finland , Estonia and Israel [30] . They included examples of all three poliovirus serotypes , although type 2 VDPVs were the most prevalent among them . These VDPV isolates showed molecular properties typical of iVDPVs described above indicating that an unknown number of these chronic excreters exist elsewhere . Interestingly , highly evolved type 1 and 2 VDPVs , which have been repeatedly isolated in Israel’s sewage in the last few years , do not appear to have spread widely as type 1 wild poliovirus did . Enhanced surveillance including sewage sampling and stool surveys should continue for as long as possible to search for the presence of iVDPV strains . The availability of efficient anti-viral treatments to interrupt virus replication in these individuals , actively being pursued at present [31] , would also be vital as previous attempts have failed [18] . These measures are needed to be able to identify and manage the possible risks of iVDPV strains spreading and causing disease in patients and the general population , particularly in the light of changes in vaccination strategies as part of the polio eradication endgame and the absence of an established outbreak response strategy . Just as the use of new monovalent and bivalent vaccines proved essential to the elimination of wild poliovirus [32 , 33] , novel vaccines unable to cause poliomyelitis would be useful at this stage of polio eradication . New polio vaccines such as those based on non-infectious virus-like particles or even new genetically designed stable live-attenuated versions [34–37] with no associated risk of producing VDPVs , might be required to resolve the “OPV paradox” that derives from using OPV to respond to outbreaks and generating new VDPVs as a consequence .
The patient is a white male from the UK . He received a full course of childhood immunisations , including OVP at 5 , 7 , and 12 months , with a booster at about 7 years of age . He was later diagnosed with common variable immunodeficiency ( CVID ) and started on intramuscular immunoglobulin therapy , which was changed to intravenous immunoglobulin after that [18] . Poliovirus was isolated from 10% stool suspensions using HEp-2c cells . Type 2 iVDPV strains 6735 and 04–44149261 , isolated from two other immunodeficient patients in the UK , and type 2 wild poliovirus strains EGY42 ( MEF-1 strain used for IPV production ) , EGY52 , VEN59 , MOR78 and KUW80 isolated from paralytic cases in 1942 , 1952 , 1959 , 1978 and 1980 , respectively , were also characterised in this study . Tg21-Bx transgenic mice expressing the human poliovirus receptor were inoculated intramuscularly ( left hind limb ) with 50 μl of 10-fold viral dilutions and daily clinical scores were recorded for 14 days . Eight mice were used for each viral dilution . The Probit method was used to calculate the 50% paralytic dose ( PD50 ) and associated 95% confidence intervals for each poliovirus challenge [27] . The antigenic properties of poliovirus isolates were studied by analysing their ability to bind Sabin 2-specific monoclonal antibodies in ELISA assays using testing formats described before [42] . Antibodies corresponding to antigenic sites 1 , 2a , 2b and 3b were used in these assays . Solutions containing equivalent concentrations of poliovirus measured as 50% cell culture infectious doses ( CCID50 ) per ml were selected . The results represent the OD values at 492nm and were expressed as normalised values relative to those obtained with antibody 1102 which reacted with all poliovirus strains . Tg21-Bx mice ( 8 per test group ) were immunised twice by intraperitoneal injection with IPV ( using the equivalent of 1 human dose/mouse ) or minimum essential medium ( diluent control ) at an interval of 2 weeks . Twenty-one days after the last dose , mice were challenged with the equivalent of 25 times the PD50 of live poliovirus and daily clinical scores were recorded for 14 days [27] . Vaccines used in these experiments were kindly donated by various manufacturers and were coded to maintain anonymity . This work was part of the characterisation of these vaccines as reference standards . It is important to note that given the high PD50 values observed for type 2 poliovirus strains in our transgenic mouse neurovirulence model , mice were challenged with very large amounts of virus ( around 108 CCID50/mouse ) . Neutralizing antibody titres in serum samples were determined by a standard microneutralization assay in 96-well plates . Two-fold serial dilutions of serum were preincubated with one hundred CCID50 of virus for 2 hours at 36°C . HEp-2C cells were added to each well , and survival at day 5 post-infection determined by staining with a 0 . 1% Naphthalene black solution . Antibody titres were expressed as reciprocals ( Log2 ) of the highest dilution of serum that protected 50% of the cell cultures determined by the Karber formula . Virus challenge doses were confirmed by back-titration . The significance of pairwise differences in neutralization titres of human sera against MEF-1 , Sabin 2 and iVDPV 160198 strain was determined using the Wilcoxon signed-rank test . The adult subject who provided stool samples gave written informed consent . Adults who provided blood samples also provided written informed consent . The work complies with the Caldicott Principles and Recommendations for patient confidentiality set up by the UK National Health Services ( NHS ) . All links to personal details that could be used to identify individuals were removed and data were analysed anonymously when possible . No samples from children were involved in the study . The study was approved by NIBSC’s Ethics and Human Materials Advisory Committees . NIBSC’s Animal Welfare and Ethical Review Body approved the application for Procedure Project Licence Number 80/2478 which was approved by the UK Government Home Office and under which animal care and protocols shown in this paper were conducted . All animal care and protocols used at NIBSC adhere to UK regulations ( Animals , scientific procedures , Act 1986 that regulates the use of animals for research in the UK ) and to European Regulations ( Directive 2010/63/Eu of the European Parliament on the protection of animals used for scientific purposes ) . The experiments in mice shown here were carried out following protocols 4 and 5 within Home Office Procedure Project Licence Number 80/2478 referred above . | The global polio eradication initiative is the most ambitious and complex public health programme directed at a single disease in history with a projected cost of $16 . 5 billion . Of the three serotypes types 2 and 3 appear to have been eradicated in the wild and type 1 is mostly confined to a region of Pakistan and Afghanistan . There is a real probability of total eradication in the near future . The main vaccine used is a live attenuated virus , and our paper concerns one of the most intractable significant implications that this has for the polio endgame . We describe virological studies of a patient deficient in humoral immunity who has been excreting type 2 vaccine-derived poliovirus for 28 years . Our results show that the viruses are excreted at high titres , extremely virulent and antigenically drifted and raise questions about how the population may best be protected from them , particularly in the light of possible changes in vaccine production which are being encouraged to increase capability and reduce costs . The study has implications for the ecology of poliovirus in the human gut and highlights the risks that such vaccine-derived isolates pose for polio re-emergence in the post-eradication era . | [
"Abstract",
"Introduction",
"Results",
"and",
"Discussion",
"Materials",
"and",
"Methods"
] | [] | 2015 | Twenty-Eight Years of Poliovirus Replication in an Immunodeficient Individual: Impact on the Global Polio Eradication Initiative |
A cellular pre-mRNA undergoes various post-transcriptional processing events , including capping , splicing and polyadenylation prior to nuclear export . Splicing is particularly important for mRNA nuclear export as two distinct multi-protein complexes , known as human TREX ( hTREX ) and the exon-junction complex ( EJC ) , are recruited to the mRNA in a splicing-dependent manner . In contrast , a number of Kaposi's sarcoma–associated herpesvirus ( KSHV ) lytic mRNAs lack introns and are exported by the virus-encoded ORF57 protein . Herein we show that ORF57 binds to intronless viral mRNAs and functions to recruit the complete hTREX complex , but not the EJC , in order assemble an export component viral ribonucleoprotein particle ( vRNP ) . The formation of this vRNP is mediated by a direct interaction between ORF57 and the hTREX export adapter protein , Aly . Aly in turn interacts directly with the DEAD-box protein UAP56 , which functions as a bridge to recruit the remaining hTREX proteins to the complex . Moreover , we show that a point mutation in ORF57 which disrupts the ORF57-Aly interaction leads to a failure in the ORF57-mediated recruitment of the entire hTREX complex to the intronless viral mRNA and inhibits the mRNAs subsequent nuclear export and virus replication . Furthermore , we have utilised a trans-dominant Aly mutant to prevent the assembly of the complete ORF57-hTREX complex; this results in a vRNP consisting of viral mRNA bound to ORF57 , Aly and the nuclear export factor , TAP . Strikingly , although both the export adapter Aly and the export factor TAP were present on the viral mRNP , a dramatic decrease in intronless viral mRNA export and virus replication was observed in the absence of the remaining hTREX components ( UAP56 and hTHO-complex ) . Together , these data provide the first direct evidence that the complete hTREX complex is essential for the export of KSHV intronless mRNAs and infectious virus production .
The nuclear export of mRNA composes one part of a larger network of molecular events that begin with transcription of the mRNA in the nucleus and end with its translation and degradation in the cytoplasm . During trafficking to the cytoplasm , a nascent mRNA undergoes numerous co-transcriptional processing steps , including 5′ capping , splicing to remove introns and 3′ polyadenylation [1]–[3] . Of these events it has become clear that splicing is particularly important for mRNA nuclear export [4] . The question of exactly which proteins regulate mRNA nuclear export has been the focus of several recent reviews [5]–[8] . Two distinct multi-protein complexes are recruited to cellular mRNAs as a consequence of splicing , namely the human transcription/export complex ( hTREX ) and the exon-junction complex ( EJC ) . The hTREX complex contains the proteins Aly ( a NXF/TAP-adapter ) , UAP56 ( a RNA-helicase ) and the hTHO-complex ( a stable complex composed of hHpr1 , hTho2 , fSAP79 , fSAP35 and fSAP24 ) [9] . A second multi-protein complex , termed the exon-junction complex ( EJC ) is deposited 20–24 nucleotides upstream of the exon-exon boundary during splicing . Until recently it was believed that Aly and UAP56 were components of the EJC [7] , [10]–[12] , however , new evidence suggests that Aly and UAP56 are associated exclusively with hTREX and not with the EJC . Therefore , these results suggest that hTREX and EJC are distinct complexes , bind at separate locations on the spliced mRNA [13] and have separate functions , where hTREX directs nuclear export of mRNA and the EJC may instead monitor mRNA fidelity and function during translation [14]–[16] . At present , it is not fully understood what regulates hTREX assembly on the mRNA but in addition to splicing the 5′ cap is also essential for its recruitment [9] , [13] . Specifically , an interaction between Aly and the cap-binding complex protein , CBP80 appears to be critical for assembly . Indeed , the 5′ cap has been shown to be required for mRNA export in Xenopus oocytes [13] . In contrast to the EJC which binds near each exon-exon boundary , hTREX is recruited exclusively to the 5′ end of the first exon , presumably regulated in part by the reported interaction between CBP80 and Aly [13] . It has been suggested that localising the export proteins at its 5′ end affords the mRNA polarity when exiting the nuclear pore . Therefore , a current model for mRNA export favours a situation where hTREX is recruited to the 5′ cap of spliced mRNA and once bound Aly stimulates the recruitment of the export factor , TAP . TAP then interacts with p15 and the nucleoporins , providing the connection between the ribonucleoprotein ( RNP ) and the nuclear pore [17] . The functional roles , if any , played by UAP56 and hTHO-complex in this process remain poorly characterised . Kaposi's sarcoma-associated herpesvirus ( KSHV ) /Human herpesvirus 8 ( HHV8 ) is a γ-2 herpesvirus associated with a number of AIDS-related malignancies including Kaposi's Sarcoma ( KS ) , primary effusion lymphoma ( PEL ) and multicentric Castleman's disease [18]–[21] . In contrast to the majority of mammalian genes , a property shared amongst all herpesviruses is that a proportion of lytically expressed viral genes lack introns . Although , KSHV expresses a higher proportion of spliced genes than other herpesviruses , it still encodes a significant proportion of lytically expressed late structural genes which lack introns . KSHV replicates in the nucleus of the host mammalian cell , and therefore requires its intronless mRNAs to be exported out of the nucleus to allow viral mRNA translation in the cytoplasm . This raises an intriguing question concerning the mechanism by which the viral intronless mRNAs are exported out of the nucleus in the absence of splicing . To circumvent this problem , and to facilitate viral mRNA export , herpesviruses of all subfamilies encode a functionally conserved phosphoprotein which has an essential role in viral lytic replication [22] . In KSHV this protein is encoded by the intron-containing open reading frame 57 ( ORF57 ) and has been the subject of several recent reviews [23]–[26] . The ORF57 gene product interacts with Aly , binds viral mRNA , shuttles between the nucleus and the cytoplasm and promotes the nuclear export of viral mRNA transcripts [27]–[31] . These properties are also conserved in ORF57 homologues such as ICP27 from Herpes simplex virus type-1 ( HSV-1 ) , SM protein from Epstein Barr virus ( EBV ) [32]–[35] and the Herpesvirus saimiri ( HVS ) ORF57 protein [27] , [31] , [36]–[38] . Here we show that KSHV ORF57 interacts during viral replication with CBP80 and hTREX , but not the EJC . We further show that ORF57 orchestrates the assembly of hTREX onto an intronless viral mRNA . The ORF57-mediated recruitment of hTREX is achieved via a direct interaction between ORF57 and Aly . Furthermore , in vitro data showed that UAP56 acts as a bridge between Aly and the hTHO-complex protein hHpr1 , thereby facilitating the formation of the complete hTREX complex . When we prevented the recruitment of Aly onto intronless viral mRNA using an ORF57 Aly-binding mutant , this resulted in a failure of ORF57-mediated viral mRNA export and significantly reduced virus replication . Strikingly , expression of a dominant negative Aly mutant that prevented the recruitment of UAP56 and hTHO-complex onto intronless viral mRNA resulted in a dramatic reduction in intronless viral mRNA export and infectious virus production . We therefore propose that the entire hTREX complex must be recruited to intronless viral mRNA by ORF57 in order for efficient intronless mRNA nuclear export and KSHV replication to occur .
The hTREX complex contains several nuclear export proteins . Given that KSHV ORF57's primary role is attributed to the nuclear export of intronless viral mRNA , we first assessed if ORF57 interacted with hTREX components using co-immunoprecipitation assays . Moreover , as hTREX forms a complex with the 5′-cap protein CBP80 [13] , we were interested if ORF57 also interacted with CBP80 . 293T cells were transfected with pGFP or pORF57GFP and untreated or RNase treated total cell lysate was used in co-immunoprecipitation experiments with CBP80- , Aly- , UAP56- , fSAP79- and hHpr1- specific antibodies in addition to an unrelated antibody control ( a p53-specific antibody ) . Each of the hTREX proteins and CBP80 co-precipitated with ORF57 , in an RNA-independent manner ( Fig . 1A ) . Moreover , indirect immunofluorescence showed that a proportion of ORF57GFP co-localised with hTREX proteins ( Fig . S1 ) . To assess whether ORF57 also interacts with the EJC , co-immunoprecipitation assays were repeated using an antibody specific for eIF4A3 , a core EJC component [39] and a hHpr1-specific antibody , serving as a positive control . No interaction was observed with the EJC core component , eIF4A3 , in contrast , ORF57 was readily detectable in the hHpr1 immunoprecipitation ( Fig . 1B ) . A control immunoprecipitation was performed to confirm that the eIF4A3 antibody precipitated EJC components ( Y14 ) in this assay ( data not shown ) . In order to address potential overexpression artefacts and to assess whether ORF57 interacts with hTREX core components during lytic replication , KSHV-latently infected BCBL-1 cells were reactivated using the phorbol-ester , TPA , and lytic gene expression confirmed by detection of the ORF57 protein in TPA-treated cells by western blot analysis ( Fig . 1C ( i ) ) . Reactivated BCBL-1 cell lysate remained untreated or was treated with RNase and co-immunoprecipitations performed using an ORF57-specific antibody . Western blot analysis using CBP80- and hHpr1- specific antibodies revealed that ORF57 interacts with CBP80 and hHpr1 during lytic replication , however ORF57 did not precipitate with either eIF4A3 ( the EJC core component ) or the cellular intronless mRNA-export protein , SRp20 ( Fig . 1Cii ) . Moreover , to confirm that ORF57 failed to interact with additional components of the EJC , co-immunoprecipitations were repeated using reactivated BCBL-1 cell lysates and Y14- and Magoh-specific antibodies . Results demonstrate that ORF57 did not precipitate with these additional EJC components ( Fig . 1Ciii ) . A control immunoprecipitation was also performed to confirm that the Y14- and Magoh-specific antibodies precipitated eIF4A3 in this assay ( Fig . S2 ) . Therefore , these data provide the first direct evidence of a viral protein associating with CBP80 and all the core components of the hTREX complex . One possible explanation for how herpesvirus intronless mRNAs undergo nuclear export is that ORF57 mimics splicing by loading key mRNA export proteins , such as hTREX , onto the intronless viral mRNA . In order to test if intronless KSHV transcripts were associated with hTREX proteins and if ORF57 was necessary for this interaction , RNA-immunoprecipitation ( RNA-IP ) assays were performed . We chose to perform this assay using 2 intronless KSHV mRNAs , specifically ORF47 and gB . RT-PCR and sequence analysis confirmed that both of these ORFs do not contain introns ( data not shown ) . To perform the RNA-IPs , a vector expressing KSHV ORF47 ( a late structural intronless gene ) was transfected into 293T cells either alone or in the presence of pORF57GFP . Total cell lysates were then used in immunoprecipitations performed with either CBP80- , Aly- , UAP56- or hHpr1-specific antibodies . RNA-IPs performed on cell extracts transfected with ORF47 alone failed to show an interaction between Aly , UAP56 or hHpr1 and the viral ORF47 mRNA ( Fig . 2A ) . In contrast , extracts from cells transfected with both pORF47 and pORF57GFP displayed a clear interaction between Aly , UAP56 and hHpr1 and the intronless viral ORF47 mRNA ( Fig . 2A ) . CBP80 was found to bind to the intronless ORF47 viral mRNA independently of ORF57 ( Fig . 2A ) . Moreover , this analysis was repeated with a second intronless KSHV mRNA , namely the late structural glycoprotein gB , and similar results were observed ( Fig . 2C ) . These data show that ORF57 is required for the recruitment of core components of hTREX onto intronless viral mRNA . To determine whether EJC components are recruited to intronless viral transcripts prior to export , RNA-IP assays were also performed using eIF4A3- , Y14- and Magoh-specific antibodies . Results failed to show any interaction between the EJC core components and viral intronless ORF47 and gB mRNAs in the absence or presence of ORF57 ( Fig . 2B and 2C ) . These results show that the EJC is not recruited to intronless viral transcripts by ORF57 and suggests that the EJC is not required for KSHV intronless viral mRNA nuclear export . To determine whether the hTREX and EJC components were recruited to a spliced viral transcript , RNA-IPs were also performed using a vector expressing the genomic ( intron-containing ) KSHV ORF50 gene . 293T cells were transfected with pORF50 in the absence or presence of ORF57 . Total cell lysates were then used in immunoprecipitations performed with either CBP80- , Aly- , UAP56- , hHpr1- , eIF4A3- , Y14- or Magoh-specific antibodies . Results demonstrated that CBP80 , hTREX and EJC components were recruited to the spliced ORF50 mRNA in an ORF57 independent manner ( Fig 2D ) . This suggests that splicing of a viral transcript is sufficient to recruit the cellular proteins necessary for nuclear export . In contrast , ORF57 is required for the recruitment of the hTREX proteins to an intronless viral transcript . Currently , while it is known that hTREX recruitment to a mammalian mRNA is both 5′-cap- and splicing-dependent , the protein-protein interactions that govern assembly of the hTREX complex itself are not fully understood . As ORF57 functions to recruit hTREX onto the intronless viral mRNA in a splicing independent manner we assessed whether this viral-system could be used to investigate hTREX assembly in more detail . To this end , we sought to determine if any hTREX proteins directly interacted with ORF57 . Radio-labelled ORF57 was generated by in vitro coupled transcription/translation ( ITT ) , RNase treated , and used in GST pull-down experiments using constructs expressing GST- , GST-Aly , GST-UAP56 and GST-hHpr1 fusion proteins . Equal amounts of each expressed protein were used in each pulldown experiment ( Fig . 3A ) . Analysis showed that ORF57 bound directly to GST-Aly but not to any other hTREX component ( Fig . 3B ) . Due to the instability of GST-CBP80 , a reverse pulldown experiment was performed using GST-ORF57 ( Fig . 3C ) and radio-labelled ITT CBP80 , a GST-Aly pulldown with ITT CBP80 served as a positive control [13] . Results also revealed a direct interaction between CBP80 and KSHV ORF57 ( Fig . 3D ) . These data suggest that ORF57 only interacts directly with Aly and CBP80 , therefore the question remains how the complete hTREX complex associates with ORF57 . It has previously been suggested that the hTREX complex is formed by UAP56 bridging the interaction between Aly and the hTHO-complex [9] . Therefore , to further investigate ORF57-hTREX assembly , we assessed which hTREX components were required to reconstitute the ORF57-hHpr1 interaction . GST pulldown experiments were performed using GST-hHpr1 and ITT ORF57 alone or combinations with ITT Aly or recombinant UAP56 . When the GST-hHpr1 ITT ORF57 pulldown was repeated in the presence of both ITT Aly and purified UAP56 , analysis revealed a clear interaction between hHpr1 and ORF57 ( Fig . 3E ) , suggesting that ORF57 requires both Aly and UAP56 to recruit the hTHO-complex , thus facilitating formation of the ORF57-hTREX complex . These findings provide the first direct evidence that UAP56 functions as a bridge between Aly and the hTHO-complex component hHpr1 to facilitate assembly of hTREX . However , at present we cannot exclude the possibility that ORF57 interacts directly with other hTHO-complex components . To assess whether hTREX is essential for viral mRNA nuclear export we produced an ORF57 mutant protein which was unable to interact with Aly and as such would be predicted to prevent the recruitment of the complete hTREX complex onto intronless viral mRNA . A minimal region responsible for Aly-binding has been identified in ORF57 and spans 35aa between residues 181 and 215 [28] . Upon closer examination of this sequence , we identified a PxxP-polyproline motif . To assess whether this motif was important for Aly-binding , both proline residues were substituted with alanine residues by site-directed mutagenesis to generate pORF57PmutGFP . To determine if mutating the PxxP-motif in ORF57 led to a loss of Aly binding , GST-Aly pulldown assays were performed using ITT ORF57 or ITT ORF57Pmut . Results demonstrated that the mutant ORF57 protein was unable to interact with GST-Aly , in contrast to the wild type protein ( Fig . 4A ) . Moreover , similar results were observed using pull-down assays with pGFP- , pORF57GFP- or pORF57PmutGFP-transfected 293T cell lysates ( Fig . 4B ) . These data demonstrate that the ORF57 PxxP-motif is required for the direct interaction with Aly . To confirm that the mutagenesis of the PxxP motif had no effect on ORF57 protein stability or other reported functions , several independent experiments were performed to assess the ability of ORF57PmutGFP to localise to nuclear speckles , homodimerise , directly interact with ORF50 and bind viral intronless mRNA ( Fig . S3 ) , all of which are features of the wild type ORF57 protein . In each case the ORF57PmutGFP phenotype was indistinguishable from that of wild type ORF57 . Having established that ORF57PmutGFP is unable to interact with Aly and that the mutation does not affect other ORF57 functions , we then asked if , in the absence of Aly-binding , ORF57 was still able to complex with CBP80 and hTREX components . 293T cells were transfected with pGFP , pORF57GFP or pORF57PmutGFP and total cell lysates were used in co-immunoprecipitation experiments , using CBP80- , Aly- , UAP56- , and hHpr1-specific antibodies . In each case the hTREX antibody immunoprecipitated ORF57GFP but not ORF57PmutGFP , demonstrating that in the absence of the Aly-interaction ORF57 was unable to form a complex with hTREX ( Fig . 4C ) . In addition , the ORF57PmutGFP exhibited a reduced but specific binding to CBP80 ( Fig . 4C ) . This reduced binding may be due to the mutation of the PxxP-polyproline motif either affecting CBP80 binding directly or the loss of hTREX binding affects the stability of the CBP80-ORF57 complex . To further investigate whether the mutation of the PxxP-polyproline motif affected direct binding to CBP80 , GST pulldown assays were performed using GST-ORF57 and GST-ORF57PmutGFP . Equal amounts of each expressed protein was incubated with radio-labelled ITT CBP80 . Results demonstrated that ORF57 and ORF57PmutGFP bound to CBP80 with similar affinity ( Fig . S4 ) . This suggests that the reduced binding observed between ORF57PmutGFP and CBP80 may be due to the loss of hTREX , which is possibly required to stabilise the export competent vRNP . To determine if ORF57PmutGFP was unable to recruit hTREX proteins to KSHV intronless mRNA transcripts in the absence of Aly binding , RNA-IP assays were performed using CBP80- , Aly- , UAP56- or hHpr1-specific antibodies . These data demonstrate that in contrast to pORF57GFP , pORF57PmutGFP is unable to recruit hTREX components to intronless viral mRNA ( Fig . 4D ) . This suggests that a direct interaction between Aly and ORF57 is required for hTREX recruitment onto intronless viral transcripts . To test if a failure in ORF57-mediated recruitment of hTREX to the intronless ORF47 mRNA prevented nuclear export of intronless KSHV transcripts , two independent mRNA export assays were performed . Firstly , northern blotting was used to detect if intronless ORF47 mRNA was present in the nuclear or cytoplasmic fraction of transfected cells . Very little ORF47 mRNA was detected in the cytoplasmic RNA fraction of cells transfected with pORF47 alone ( 9 . 9±4 . 9% ) , whereas cells co-transfected with pORF47 and pORF57GFP displayed a clear shift in ORF47 mRNA from the nuclear to the cytoplasmic fraction ( 81 . 5±1 . 0% ) , indicative of ORF57-mediated viral mRNA nuclear export . However , upon co-transfection with pORF47 and pORF57PmutGFP , the majority of ORF47 mRNA was no longer found in the cytoplasmic fraction ( 21 . 3±3 . 8% ) , instead it was retained in the nuclear pool at similar levels to those seen for the negative control , symptomatic of a failure in ORF57-mediated viral mRNA nuclear export ( Fig . 5A ) . To confirm that the ORF57 mutant did not affect mRNA stability , total RNA levels were assessed by northern blot analysis . No significant difference in ORF47 mRNA levels was observed between cells expressing wild type or mutant ORF57 proteins ( Fig . 5A , right panel ) . However , a slight decrease in total mRNA levels is seen in the presence of both the ORF57 or ORF57PmutGFP compared to the GFP control . At present , the reason for this is unknown , however , it could be due to the overexpression of the ORF57 protein . To confirm the above result , a fluorescent in situ hybridisation assay was utilised . 293T cells were transfected with pORF47 , in addition to either pGFP , pORF57GFP or pORF57PmutGFP . 24 h post-transfection cells were fixed , permeabilised and incubated with a biotin-labelled oligonucleotide specific for the KSHV ORF47 mRNA . After a 4 hr hybridisation cells were washed and ORF47 mRNA subcellular localisation was visualised using Cy5-streptavidin . Cells transfected with pORF47 and GFP retained the ORF47 mRNA in the nucleus , whereas ORF47 mRNA was clearly visualised in the cytoplasm of cells transfected with pORF47 and pORF57GFP . However , upon transfection with pORF57PmutGFP , ORF47 mRNA was only observed in the nucleus , symptomatic of a failure in ORF57-mediated viral mRNA nuclear export ( Fig . 5B ) . Together , these two independent assays demonstrate that the ORF57-dependent recruitment of hTREX to intronless viral transcripts is essential for their efficient nuclear export . We were also interested to determine whether the recruitment of the complete hTREX complex is required for virus replication and infectious virion production . To this end , we utilised a 293T cell line harbouring a recombinant KSHV BAC36-GFP genome [40] . This KSHV-latently infected cell line can be reactivated releasing infectious virus particles in the supernatant which can subsequently be harvested and used to infect 293T cells [41] . The 293T-BAC36 cell line was transfected with pGFP , pORF57GFP or pORF57PmutGFP and concurrently reactivated using TPA and incubated for 72 hours . The supernatants from each flask were then harvested and used to re-infect 293T cells and GFP positive cells were scored 48 h post-infection , as described above . Results revealed similar levels of lytic replication and virus production from cells expressing pGFP or pORF57GFP . However , virus production was significant reduced ( P = 0 . 018 ) upon the expression of the ORF57PmutGFP ( Fig . S5 ) . Therefore , these results demonstrate that the ORF57-dependent recruitment of the complete hTREX complex to intronless viral transcripts is essential for efficient virus lytic replication and infectious virion production . The above data show that ORF57 binds viral intronless mRNA and directly interacts with Aly . Given that Aly is able to recruit the export factor TAP directly , it was of interest to determine if UAP56 and the hTHO-complex are required for viral mRNA export . In contrast to the cellular mRNA model , a major advantage of our viral system is that hTREX assembly on the viral mRNA is dependent upon an interaction with a virus-encoded protein , not splicing . Specifically , ORF57 binds viral mRNA , directly interacts with and recruits Aly which in turn then interacts with and uses UAP56 to bridge an interaction with the hTHO-complex . This ordered recruitment allows us to specifically disrupt the viral mRNA-ORF57-hTREX complex at different points and assess the functional significance on nuclear export . Furthermore , rather than using an artificial in vitro assay to investigate the functional significance of hTREX , we assessed this in the context of the virus replication cycle using the 293T-BAC36 assay described above . The trans-dominant mutant , pAlyΔC-myc , which has 20 residues deleted from the carboxy-terminus of Aly , is unable to interact with UAP56 [42] . We were interested in establishing if this mutant could be used to disrupt the assembly of UAP56 and hTHO-complex on an intronless viral mRNA and as such provide insights into whether these proteins are essential for nuclear export . However , prior to its use in the replication assay it was essential to confirm that AlyΔC-myc is still recruited by ORF57 to intronless viral mRNA and is able to interact with TAP . To this end , ORF57 , UAP56 and TAP were expressed as GST fusion proteins and incubated with either pmyc , pAly-myc or pAlyΔC-myc transfected cell lysates and pulldown analysis performed . Western blotting using a myc-specific antibody demonstrated that Aly-myc interacted with ORF57 , TAP and UAP56 . In contrast , AlyΔC-myc is unable to associate with UAP56 but retains the ability to interact with both ORF57 and TAP ( Fig . 6A ) . These results suggest that AlyΔC-myc is an ideal mutant to inhibit the recruitment of UAP56 and hTHO-complex on the viral intronless mRNA . However , one caveat to this system is that expression of pAlyΔC-myc may also act in a dominant negative capacity to inhibit spliced mRNA nuclear export [42] . Therefore it was important to allow expression of the spliced ORF57 protein prior to accumulation of pAlyΔC-myc . To this end , transient transfection of pAlyΔC-myc was performed concurrent with reactivation of the KSHV lytic replication cycle , and ORF57 protein levels assessed 24 h later . Results show that comparable amounts of ORF57 were expressed in untransfected , pmyc , pAly-myc and pAlyΔC-myc transfected cell lysates ( Fig . 6B ) . To test if AlyΔC-myc inhibited the recruitment of UAP56 and the hTHO-complex onto KSHV intronless RNAs , RNA-IPs were performed on reactivated KSHV-infected 293T cells transfected with pmyc , pAly-myc or pAlyΔC-myc . We obtained similar results for pmyc and pAly-myc to those shown in Fig . 2A , where recruitment of hTREX components onto the viral RNA was readily detected 48 h-post reactivation . However , RNA-IPs using cell extracts transfected with AlyΔC-myc showed a dramatic decrease in the recruitment of UAP56 and hHpr1 to viral mRNA ( Fig . 6C ) . RNA-IPs performed using a TAP-specific antibody showed that TAP is recruited to the intronless viral mRNA , irrespective of Aly status . Critically , RNA-IPs using an ORF57-specific antibody produced ORF47 RT-PCR products of a similar intensity , suggesting that ORF57 was not limiting in this assay ( Fig . 6C ) . It should also be noted that we observed a decrease in TAP recruitment to the viral mRNA in the presence of both pAly-myc and pAlyΔC-myc , compared to pmyc control . At present , we are unsure why TAP recruitment is reduced , however , no difference in mRNA nuclear export is observed between pmyc and pAly-myc transfected cells , suggesting that this reduction in TAP recruitment does not impede the nuclear export of intronless viral mRNAs . To assess if the AlyΔC-myc mutant affected intronless viral mRNA export during replication , northern blot analysis was performed as described above . Results demonstrated that ORF47 mRNA nuclear export is impaired in reactivated cells that expressed AlyΔC-myc , but not in cells expressing myc or Aly-myc ( Fig . 6D ) . Moreover , to determine if expression of AlyΔC-myc had any effect on virus replication , the KSHV-latently infected 293T BAC36-GFP cell line was transfected with pmyc , pAly-myc or pAlyΔC-myc and concurrently reactivated using TPA and incubated for 72 hours . The supernatants from each flask were then harvested and used to re-infect 293T cells . The level of virus replication was determined by scoring the percentage of GFP positive cells 48 h post-infection , as previously described [41] . Similar levels of lytic replication and virus production were observed from pmyc and pAly-myc pre-transfected cells . Strikingly , virus production from pAlyΔC-myc pre-transfected cells was reduced by approximately 10 fold ( Fig . 6E ) . These data demonstrate that ORF57-mediated recruitment of Aly and TAP to an intronless viral mRNA is insufficient for its nuclear export and that a lack of UAP56 and hTHO-complex on an intronless viral mRNA has a profound effect on intronless nuclear export and KSHV lytic replication .
Co-immunoprecipitation data show that ORF57 readily associates with components of hTREX , however , no such interaction was observed between ORF57 and the EJC proteins; eIF4A3 , Y14 and Magoh . This result suggests that the EJC is not recruited to intronless viral transcripts and this was confirmed using RNA-IP assays . In contrast , hTREX proteins readily precipitated with intronless viral mRNA , in the presence of ORF57 , which presumably functions as a linker between hTREX and the viral mRNA . These findings suggest that the essential export adapter complex for intronless KSHV nuclear export is hTREX and not the EJC . It should be noted that these findings are in contrast to previous observations using a homologue of KSHV ORF57 from the prototype γ-2 herpesvirus , Herpesvirus saimiri [29] . One possible explanation for these contrasting data is that co-immunoprecipitations from Williams et al . were performed by over-expressing myc-tagged EJC components , whereas this in study , endogenous EJC proteins was precipitated using an eIF4A3- , Y14 and Magoh-specific antibodies . To test this , we have performed co-immunopreciptations with EJC specific-antibodies using HVS-infected cell lysates . No interactions were observed between HVS ORF57 and the endogenous EJC proteins ( Fig . S6 ) , suggesting the previously observed interactions may have been due to the overexpression of the EJC components . In addition to splicing dependency , the cap-binding complex protein , CBP80 , is required to recruit hTREX to human pre-mRNA , via a direct interaction with Aly . Interestingly , we detected a direct interaction between ORF57 and CBP80 , implying that the 5′ cap may also function in intronless KSHV mRNA export . However , upon disrupting the ORF57 and Aly interaction ( via mutation of the PxxP motif ) , we also observed a reduction of the ORF57-CBP80 interaction . Analysis suggests that this reduction maybe due to the loss of hTREX affecting the stability of the export competent viral RNP . This suggests that although ORF57 interacts directly with Aly and CBP80 , these interactions may not overlap and more detailed analysis of the interacting domains for both proteins is required . It is also worth noting however , that in the absence of ORF57 , CBP80 did not recruit Aly to the intronless viral transcripts , suggesting that ORF57 is essential for the loading of hTREX on viral mRNA . The lack of EJC recruitment to intronless viral mRNA may have ramifications beyond those of nuclear export , for example , the EJC has been suggested to function in translational efficiency [16] , [43] . Intriguingly , the herpes simplex virus type-1 ( HSV-1 ) ORF57 homologue , ICP27 , has been implicated in increased translation efficiency [44] , [45] , we are currently investigating whether ORF57 increases translation of KSHV transcripts during virus replication . The current model for hTREX assembly on a spliced mRNA describes UAP56 and Aly associating with the mRNA in a 5′ cap- and splicing-dependent manner . Moreover , as shown in Fig . 7A , it has been suggested that UAP56 may bridge an interaction between Aly and the hTHO-complex [9] , [42] . In contrast , during KSHV replication hTREX appears to be tethered to an intronless KSHV mRNA via an exclusive interaction with ORF57 . Taking advantage of this , we used the ORF57-hTREX complex to gain insight into how individual components of hTREX interact with one another . Our data show that ORF57 interacts exclusively with Aly , which then binds directly to UAP56 and this in turn functions as a bridge to recruit hHpr1 and presumably the complete hTHO-complex ( Fig . 7B ) . This order of hTREX assembly is in broad agreement with the model proposed by Cheng et al who showed using RNase H digestion analysis that Aly was the most 5′ of the hTREX components , with UAP56 and hTHO-complex binding further downstream . Interestingly , the direct interaction observed between ORF57 and CBP80 suggest that ORF57 may recruit hTREX to the 5′ end of the intronless mRNA , perhaps to provide directionality to nuclear export as is the proposed case for spliced human mRNA [13] . The functional significance of hTREX recruitment to intronless viral mRNA is substantiated using an ORF57 point mutant and a dominant-negative Aly mutant . Specifically , we were able to disrupt the direct interaction between ORF57 and Aly by mutating two proline residues within a region of ORF57's Aly-binding domain [28] . This ORF57Pmut was still able to recognise and bind intronless viral mRNA , however , it lacked the ability to recruit hTREX to these transcripts . A failure to recruit hTREX rendered the ORF57Pmut non-functional as a viral mRNA export protein and provides direct evidence that the hTREX complex is essential for the efficient export of intronless viral mRNA and virus replication . The export adapter Aly is able to interact directly with the export factor complex TAP/p15 [46] , therefore , we were interested in assessing whether Aly-TAP/p15 recruitment produced an export-competent intronless viral mRNP or if UAP56 and the hTHO-complex were also required for nuclear export . This is of particular importance as a number of ORF57 homologues , such as Herpes simplex virus-type 1 ICP27 , have been shown to interact with Aly and TAP , but it is unknown whether they also recruit UAP56 and the hTHO-complex [32] , [33] . One major advantage of using the KSHV system to study hTREX assembly in contrast to analysing hTREX recruitment to a spliced human mRNA is that recruitment of hTREX on an intronless viral transcript is mediated via a direct interaction between ORF57 and Aly , which serves to target the remainder of hTREX to the intronless viral mRNA . This facilitated the use of a trans-dominant Aly mutant , termed AlyΔC , which retains a direct interaction with ORF57 yet fails to interact with UAP56 . The AlyΔC mutant has limited use as a tool for dissecting hTREX recruitment to spliced human mRNA as it does not bind to spliced mRNA [42] . The introduction of the dominant-negative AlyΔC mutant into a KSHV virus replication system dramatically reduced the amount of UAP56 and hHpr1 recruited to intronless viral transcripts and this in turn led to a striking reduction in intronless viral mRNA nuclear export and significantly , virus replication . Importantly , RNA-IP analysis of intronless viral mRNPs from cells expressing AlyΔC revealed that ORF57 , Aly and TAP where all present on intronless viral mRNA , suggesting that UAP56 and perhaps the hTHO-complex possess an unidentified , yet essential role in mRNA nuclear export . These data place hTREX at the hub of human mRNA nuclear export . However , RNAi studies in Drosophila melanogaster and Caenorhabditis elegans have shown Aly to be non-essential for mRNA export in these systems [47] , [48] . In addition , a genome-wide RNAi study in D . melanogaster reported that the conserved THO-complex was only required by a subset of transcripts for nuclear export [49] . Interestingly , D . melanogaster and C . elegans do require UAP56 both for viability and for bulk mRNA nuclear export [10] , [50] . This suggests that while there may be redundancy in eukaryotic systems for certain TREX components , others remain essential . Similar controversy surrounds the role of Aly in herpesvirus mRNA export . In contrast to our data , which show the ORF57-Aly interaction to be essential for efficient intronless viral mRNA nuclear export , a study reported that depletion of Aly using RNAi had little effect on ORF57-mediated transactivation [51] . In addition , a second manuscript reports that differences in the ORF57-Aly binding affinity does not effect ORF57 export function [52] . One possible explanation for these discrepancies is that only partial depletion of Aly was achieved by RNAi , and that such a small reduction in total Aly protein ( less than 25% compared to control ) may not be functionally significant . Likewise , the mutant ORF57 proteins described by Nekorchuk et al ( 2007 ) also failed to interact with viral mRNA , which makes it difficult to interpret their significance with regards to ORF57-mediated nuclear export of viral mRNA . Our findings using a naturally occurring intronless viral mRNA may provide some insight to the nuclear export of cellular intronless mRNAs , which are often studied using in vitro-transcribed cDNAs . The H2A intronless mRNA is exported by SRp20/9G8 that recognise and bind a specific sequence in the target intronless mRNA and subsequently promote export via a direct interaction with TAP ( Fig . 7C i ) [53] , [54] . Our data suggest that ORF57 may function in a similar manner to cellular SR-proteins by binding to a sequence-specific region of the intronless viral mRNA . Work is currently underway in our laboratory to identify potential ORF57-target sequences in intronless viral mRNA . Conversely , it will be of interest to determine whether hTREX is recruited to H2A mRNA by SR-proteins . More recently , Aly was shown to be recruited to an in vitro transcribed intronless β-globin construct independently of splicing , via a direct interaction with the CBC protein , CBP20 [55] . In contrast , a second publication suggested that Aly is recruited to in vitro transcribed intronless mRNA by UAP56 in ATP-dependent manner [56] . Interestingly , while there are some disparities , the observations made by both groups generally support a model whereby the 5′ cap , Aly and UAP56 are involved in intronless mammalian mRNA nuclear export ( Fig . 7C ii and iii ) . Once again , it will be interesting to see if further analysis reveals the presence of an entire hTREX complex on these mRNAs . In summary , these data highlight that a complete hTREX complex is required for efficient KSHV intronless mRNA export and replication . Importantly , data herein demonstrate that recruitment of the nuclear export factor TAP and its adapter protein , Aly , are not sufficient to promote nuclear export . These data suggests that UAP56 and the hTHO-complex must be recruited in order to form an export competent KSHV intronless mRNP .
Oligonucleotides used in cloning , RT-PCR analysis and mutagenesis can be found in Table S1 . To generate pORF57GFP , ORF57 cDNA was amplified by PCR and cloned into pEGFP-N1 ( BD Biosciences Ltd ) . pORF57GFPpmut was generated using the QuickChange II site-directed mutagenesis kit ( Stratagene ) . pORF47 and pORFgB were cloned into pCDNA3 . 1+ ( Invitrogen ) . pETORF57 and pETORF57pmut were cloned in pET21b ( Novogen ) . The genomic ORF50 gene was cloned into pCS2MT+ [57] . To generate pGST-ORF57pmut , the NcoI/HindIII fragment of pGST-ORF57 [28] was replaced with the NcoI/HindIII fragment from pORF57GFPpmut . pAlymyc and pAlyΔmyc [42] , pGST-Aly , pGST-UAP56 , pGST-hHpr1 and pUAP56-His [9] and pET-CBP80 [13] have all been described elsewhere . SRp20 , Y14 , and Magoh ( Santa Cruz Biotech ) , p53 ( Pharmagen Inc ) , GFP mAb and GFP pAb ( BD Biosciences ) and Myc , SC-35 , B-actin and GAPDH ( Sigma ) were purchased from the respective suppliers . Specific antibodies to CBP80 , Aly , UAP56 , hHpr1 , fSAP79 , hTho2 and eIF4A3 were previously described [9] , [13] . Unless stated all antibodies were used at a dilution of 1:1000 for western blot analysis . HEK-293T , HeLa cells and 293T BAC36 cells harbouring a recombinant KSHV BAC36 genome [41] were cultured in Dulbecco's modified Eagle medium ( DMEM , Invitrogen , Paisley , UK ) supplemented with 10% foetal calf serum ( FCS , Invitrogen ) , glutamine and penicillin-streptomycin . KSHV-infected BCBL-1 cells were cultured in RPMI medium ( Invitrogen , Paisley , UK ) supplemented with 10% foetal calf serum ( FCS , Invitrogen ) , glutamine and penicillin-streptomycin . 293T BAC36 cells and BCBL-1 cells were reactivated using TPA ( 20 ng/ml ) for 24 h . Plasmid transfections were carried out using Lipofectamine™ 2000 ( Invitrogen , Paisley , UK ) , as per the manufacturer's instructions . Glutathione S-transferase ( GST ) pulls downs and co-immunoprecipitations , in addition to subsequent protein analysis by SDS-PAGE and western blot , were performed as previously described [29] . Confirmation of successful RNase was carried out as described by Carlile et al ( Fig . S7 ) [55] , [58] . A polyclonal antibody to p53 served as an unrelated antibody control throughout . This antibody precipitated the cognate p53 protein ( Fig . S8 ) . RNA-immunoprecipitations were performed as previously described [59] . Nuclear and cytoplasmic RNA for northern blots was extracted from transiently transfected cells using the PARIS™ kit ( Ambion Inc . , Warrington , UK ) as per the manufacturer's instructions . Northern blots were carried out as described previously [60] . Membrane bound RNA was hybridised with 32P-radiolabelled random-primed probes specific for ORF47 and 18S rRNA . The blots were then analysed using a FUJIX BAS1000 Bio-Imaging Analyser ( Fuji Photo Film Co . Ltd ) and data quantified using the AIDA ( Advanced Image Data Analyser ) version 2 . 31 software . In situ hydridisation was performed as previously described [61] . 25 ng of biotin-labelled probes specific for KSHV ORF47 mRNA were denatured and hybridised at 37°C for 4 hrs . For detection , cells were incubated for 30 min with 7 µl of 12 . 5 µg/ml of Cy5-streptavidin ( Molecular probes ) . Coverslips were mounted in Vectorshield® mounting medium ( Vector Laboratories , CA ) and staining visualised on an Upright LSM 510 META Axioplan 2 confocal microscope ( Zeiss ) using the LSM Imaging software ( Zeiss ) . 293-T BAC36 cells harbouring KSHV BAC36 under hygromycin selection were reactivated using 20 ng/ml TPA . At 72 h post reactivation , filtered tissue culture supernatants were used to spinoculate 1×105 HEK 293-T cells in the presence of 5 µg/ml polybrene . Infected EGFP-positive cells were quantified at 48 h post-infection by fluorescence microscopy . | Following gene expression in the nucleus , newly transcribed messenger RNA ( mRNA ) is exported to the cytoplasm , where it is translated into protein . In mammals the vast majority of mRNAs contain introns that must be removed by the spliceosome prior to nuclear export . In addition to excising introns , splicing is also essential for the recruitment of a several protein complexes to mRNA , one example being the human transcription/export complex , which is required for mRNA export . Herpesviruses , such as Kaposi's sarcoma–associated herpesvirus , replicate by hijacking components of the host cells biological machinery , including those proteins necessary for mRNA export . An intriguing caveat in herpesvirology is that herpesviruses , such as Kaposi's sarcoma–associated herpesvirus , produce some mRNAs that lack introns and do not undergo splicing . How then are these intronless mRNAs exported to the cytoplasm ? The answer lies in a virus protein called ORF57 that is able to bind to the intronless mRNA and then export them to the cytoplasm . ORF57 achieves this function by mimicking splicing and recruiting the human transcription/export complex to the intronless viral mRNA , thus facilitating its export into the cytoplasm . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"molecular",
"biology/mrna",
"transport",
"and",
"localization",
"virology/viral",
"replication",
"and",
"gene",
"regulation",
"virology"
] | 2008 | Recruitment of the Complete hTREX Complex Is Required for Kaposi's Sarcoma–Associated Herpesvirus Intronless mRNA Nuclear Export and Virus Replication |
Plasmacytoid Dendritic Cells ( pDCs ) represent a key immune cell in the defense against viruses . Through pattern recognition receptors ( PRRs ) , these cells detect viral pathogen associated molecular patterns ( PAMPs ) and initiate an Interferon ( IFN ) response . pDCs produce the antiviral IFNs including the well-studied Type I and the more recently described Type III . Recent genome wide association studies ( GWAS ) have implicated Type III IFNs in HCV clearance . We examined the IFN response induced in a pDC cell line and ex vivo human pDCs by a region of the HCV genome referred to as the HCV PAMP . This RNA has been shown previously to be immunogenic in hepatocytes , whereas the conserved X-region RNA is not . We show that in response to the HCV PAMP , pDC-GEN2 . 2 cells upregulate and secrete Type III ( in addition to Type I ) IFNs and upregulate PRR genes and proteins . We also demonstrate that the recognition of this RNA is dependent on RIG-I-like Receptors ( RLRs ) and Toll-like Receptors ( TLRs ) , challenging the dogma that RLRs are dispensable in pDCs . The IFNs produced by these cells in response to the HCV PAMP also control HCV replication in vitro . These data are recapitulated in ex vivo pDCs isolated from healthy donors . Together , our data shows that pDCs respond robustly to HCV RNA to make Type III Interferons that control viral replication . This may represent a novel therapeutic strategy for the treatment of HCV .
Pathogens are sensed by host pattern recognition receptors ( PRR ) that recognize molecular motifs . Two major receptor systems sense the presence of viral infection to mount an immune response: toll-like receptors ( TLRs ) 3 , 7 , 8 , and 9 are major PRRs that respond to different types of viral nucleic acids , and more recently , retinoic acid inducible gene-I ( RIG-I ) -like receptors ( RLRs ) , helicases including RIG-I and MDA-5 ( melanoma differentiation-associated gene 5 ) , have been identified as cytosolic receptors for intracellular dsRNA sensing [1] , [2] . The relative contributions of TLRs and RLRs as viral sensors vary according to viruses and across different cell types [1] . By specializing in the production of Type I Interferons ( IFNs ) , i . e . IFN-α and IFN-β , plasmacytoid DCs ( pDCs ) play crucial roles as mediators of antiviral responses [3] , [4] . RIG-I signaling has been described as largely dispensable for pDC secretion of IFN-α following infection with RNA viruses [1] , whereas the TLR system is critical for the RNA virus-mediated IFN response in pDC [5] . Affecting an estimated 200 million people globally , hepatitis C virus ( HCV ) is the world's most common blood-borne viral infection for which there is no vaccine [6] . The majority of individuals exposed to this RNA virus will develop viral persistence; however , there are significant differences in how patients respond to HCV infection and its treatment [7] . HCV infection has been associated with depletion and functional suppression of pDCs [8] , [9] . Some studies have shown that pDCs from the blood of patients with chronic HCV are infected [10] , whereas others have failed to demonstrate viral infection within the pDCs . Moreover , some viruses can activate pDCs to produce Type I IFN without the need for active replication [4] , [11] . Direct contact with HCV-infected hepatocytes induces Type I IFN via TLR7 signaling within pDCs [12] . Furthermore , Chisari's group has demonstrated that HCV RNA is transferred to the pDCS from hepatocytes via a non-endocytic mechanism [13] . Although their receptor subunits do not display any detectable homology , Type III IFNs are functionally similar to Type I IFNs , signaling through JAK-STAT intracellular pathways and upregulating the transcription of IFN-stimulated genes ( ISGs ) required to control viral infection [7] . Recently , considerable data have linked genetic variation within or near the Type III IFN-λ3 ( IL-28B ) gene with HCV recovery [14] , [15] . In this study , for the first time , we define the consequences of human pDC sensing of the HCV genome 3′ non-translated poly-U/UC tract , previously shown to function as the HCV pathogen-associated molecular pattern ( PAMP ) [16] substrate of cytosolic RIG-I . Extensive analysis and characterization of the HCV genomic RNA has identified that the pU/UC tract in the 3′ UTR had the greatest capacity to stimulate IFN-β production in hepatocytes [16] . We hypothesized and now show that the HCV PAMP triggers robust and varied IFN responses from the human pDC cell line GEN2 . 2 and the secreted Type III IFNs inhibit viral replication . We demonstrate that neutralization of Type III IFNs ( IFN- λ ) attenuates the anti-HCV effects of pDC-GEN2 . 2-derived supernatants . Addition of pDC-derived supernatants activates the JAK/STAT pathway in the Huh7 . 5 . 1 hepatoma cell line . Furthermore , we demonstrate that laboratory-determined concentrations of Type III IFNs inhibit viral replication . Finally , we recapitulate the induction of Type III Interferons following recognition of the pU/UC RNA in freshly-isolated ex vivo human pDCs and show their capacity to inhibit replication .
Due to the extremely low frequency of pDCs circulating in normal healthy adults [8] , we used the GEN2 . 2 pDC cell line ( pDC-GEN2 . 2 ) [17] that phenotypically expresses classic markers of pDCs , as well as various other surface markers relevant to co-stimulation ( Figure 1 ) . Because our focus was to understand HCV pathobiology , we cultured the pDC-GEN2 . 2 using the human hepatoma cell line Huh7 . 5 . 1 ( known to support HCV replication ) as the feeder cells . For all of our single-cell experiments using the pDC-GEN2 . 2 cell line , we used the non-adherent fraction of the cultures . Using only the non-adherent fraction from these cultures , we determined that the cells were over 95% BDCA-2+ and CD45+ thus indicating that the feeder line did not contaminate the cell fraction used for experiments . We felt that using a human hepatic cell line as feeders would make our system more relevant to the study of HCV . The cells were positive for Blood Dendritic Cell Antigen ( BDCA ) -2 ( Cluster of Differentiation [CD] 303 ) , HLA-DR and CD123 , markers of pDCs , but were negative for CD11c , BDCA-1 and BDCA-3 , typical markers of myeloid DCs . Of the non-adherent cells , over 95% were BDCA-2 , HLA-DR double positive . While there is the possibility of contamination from the adherent Huh7 . 5 . 1 feeder cells , we feel that this contamination would be minor and not affect the overall results of our studies . Additional characterization of the non-adherent cells showed that these cells expressed CD86 , CD44 and CD119 ( IFNγR ) with low levels of CD80 , CD83 , CD209 ( DC-SIGN ) and IFNαR1 . TLR2 and TLR7 mRNA was consistently detected , and TLR3 and TLR9 protein levels were demonstrable by flow cytometry . The cells showed a round , even morphology as expected from suspended dendritic cells ( Figure 1D ) . The absence of other morphologies within our microscopic samples provides evidence that the non-adherent fraction of our cultures consist primarily of the pDC-GEN2 . 2 cells . Additionally , we genotyped the cell line for an IL-28B SNP ( IFN- λ3; rs12979860 ) that is associated with HCV clearance [14] , [15] and a RIG-I SNP ( rs10813831 ) that has been associated with differential trends in IFNβ1 production [18] . In the case of the IL-28B SNP , the genotype associated with HCV clearance is the CC homozygous allele [14] , [15] and was identified as the genotype present in the pDC-GEN2 . 2 cell line . The RIG-I SNP is an amino acid change ( genotype G to A; Arg to Cys ) in the protein sequence of RIG-I [18] . Patients with the rare homozygous AA genotype demonstrated a trend of higher RIG-I and IFNβ1 expression levels [18] . The pDC-GEN2 . 2 cell line was determined to have the AA genotype for this RIG-I SNP . Given the characterization , we concluded that these cells were a reasonable proxy of pDCs for our in vitro experiments . We investigated the profile of IFN transcripts following stimulation with viral TLR ligands in vitro . pDCs recognize RNA and DNA viruses through two endosomal sensors , TLR 7 and TLR9 , respectively , that induce Type I IFN secretion through the Myeloid differentiation primary response gene ( MyD88 ) -Interferon regulatory factor 7 ( IRF7 ) signaling pathway [4] . By examining the mRNA levels of IFN genes after 6 hours of co-culture with poly I:C ( TLR3 ) , Loxiribine ( TLR7/8 ) , and Oligodeoxynucleotide ( ODN ) 2216 ( Type A CpG molecule , TLR9 ligand ) , we characterized ability of the pDC-GEN2 . 2 cell line to respond to a variety of viral stimulants . As shown in Figure 2 , there was significant induction of IFNα1 , IFNα2 , IFNβ1 , IL-28A ( IFNλ2 ) , and IL-29 ( IFNλ1 ) . Not surprisingly , combined TLR3 and TLR7/8 ligation induced the most robust induction of interferon genes given that transcripts for TLR7 were consistently present . Additionally , we found up-regulation of RIG-I , TLR2 and TLR8 mRNA within the pDC-GEN2 . 2s . Next , we wanted to determine whether a feedback loop of Type I IFN contributed to these IFN expression patterns . To understand how these cells respond to IFNα2 stimulation , we co-cultured the cells with pegylated-Interferon-α2 ( commonly used in HCV therapy [19] along with the purine analog ribavirin ) for 6 hours and then examined the IFN responses by qRT-PCR . Compared to stimulation with viral TLR ligands , Type I and III IFN showed relatively diminished or absent transcriptional effect , indicating that the response seen after stimulation with viral TLR ligands is not solely attributable to an IFNα2 feedback mechanism ( Figure S1 ) . The poly-U/UC ( pU/UC ) tract of the HCV genome 3′ non-translated region functions as the PAMP substrate of retinoic acid-inducible gene ( RIG-I ) , the cytosolic PRR for HCV [16] , [20] . Of the several regions that were tested in the Saito et al . study , this pU/UC region was found to be the most stimulatory region [16] . The nearby X-region , a highly conserved region in the 3′ untranslated region ( UTR ) of the HCV genome , is not immunogenic and thus was used as a negative control ( Figure 3A ) in these experiments . Transfection of the pU/UC RNA into the pDC-GEN2 . 2 cells , in the absence of the feeder cells , lead to pronounced transcriptional induction of multiple IFNs ( Figure 3B ) , beginning as early as 2 hrs and in general , peaking at 8 hrs , but remaining elevated after 24 hrs of transfection . The X-region RNA did not consistently induce greater IFN production compared to mock transfection ( Figure S2B ) . In comparison to the induction of IFN genes by pU/UC RNA , the IFN gene expression induced by the X-region RNA was relatively weak ( Figure S2 ) . The 2′-5′ oligoadenylate synthetases ( OAS ) are a family of antiviral proteins that are induced by both virus infection and IFN stimulation and activate latent endoribonuclease ( RNase L ) [21] , yielding RNA cleavage products that initiate innate signaling [22] . HCV pU/UC RNA transfection increased OAS-1 transcription in pDC-GEN2 . 2 cells more than 40-fold at 8 hrs ( Figure 3C ) . However , RNaseL , constitutively expressed in lymphatic tissue [23] , was not further induced or upregulated by HCV PAMP transfection ( Figure S3 ) . Furthermore , transfection with the HCV PAMP upregulated transcriptional expression of PRR signaling genes , in particular RIG-I ( Figure 3C ) . Supernatants from pDC-GEN2 . 2 cultures that were stimulated with pU/UC RNA , X-region RNA or mock transfection were assayed for levels of IFN proteins by ELISA ( Figure 3 D–G ) . At 24 hours post transfection , high levels of all IFN proteins assayed were observed in the pU/UC stimulated condition . The pU/UC IFN protein levels were significantly higher than X-region IFN protein levels for both Type I and III IFNs , in accordance with the gene expression data ( Figure 3B ) . The average concentrations of IFNα , IFNβ , IL-28A ( IFNλ2 ) and IL-29 ( IFNλ1 ) were 2600 pg/mL , 650 pg/mL , 1500 pg/mL and 475 pg/mL , respectively . IFNα , IFNβ and IL-29 ( IFNλ1 ) were also detectable at 8 hours post-transfection ( data not shown ) . IL-28B ( IFNλ3 ) could not be assayed by ELISA so we examined IL-28B ( IFNλ3 ) protein levels in cell lysates from the stimulated pDC-GEN2 . 2 cells by Western blot from three independent experiments . We found that IL-28B ( IFNλ3 ) was detectable in the cell lysates of these cells and that more IL-28B was present in the pU/UC condition than either the mock or X-region condition ( Figure 3H ) although this difference was not statistically significant . PRR protein levels were also increased after 8 and 24 hours of pU/UC stimulation compared to mock or X-region transfection ( Figure 3I , left ) . Densitometry revealed a significant increase in protein level for RIG-I and MDA-5 after 24 hours in the pU/UC RNA transfection compared to the X-region RNA transfection ( Figure 3I , center ) . MyD88 and TRIF protein levels were not increased . The increases in protein are consistent with the gene expression data from 8 hours where each gene showed a significant upregulation from pU/UC stimulation compared to the X-region stimulation with the most robust upregulation in RIG-I and MDA-5 ( Figure 3I , right ) . Together , these data demonstrate that pDC-GEN2 . 2 cells not only respond to HCV RNA with IFN gene expression and protein production , but the cells also upregulate PRR genes and proteins , particularly RIG-I which is the known cytosolic receptor for the HCV PAMP . It has been shown previously that RIG-I requires 5′ triphosphate groups for recognition and signaling [24] . TLR3 and TLR7 have not been shown to have the same requirement for this biochemical feature [25] . We hypothesized that if RIG-I were involved in the recognition and response to the HCV PAMP RNA , removal of the 5′ phosphate groups would abrogate the production of IFN mRNA . We treated the in vitro transcribed HCV RNA with Antarctic phosphatases and then transfected 1 µg of the treated RNA into pDC-GEN2 . 2 cells . After treatment with the phosphatases , much of the IFN mRNA production was lost . This was true of Type I and Type III IFN message ( Figure 4A ) . Additionally , siRNA knockdown of RIG-I in the GEN2 . 2-pDCs resulted in significantly decreased Type I and Type III IFN message after stimulation with the pU/UC RNA compared to the scrambled siRNA condition ( Figure 4B ) . These findings suggest that although TLRs have been considered the primary mode of recognition in pDCs , RIG-I is involved in the recognition of the HCV PAMP RNA . While much of the IFN mRNA production was lost , there was still some production of IFN mRNA , suggesting that in addition to RIG-I , other PRRs might be involved in the recognition of this RNA . In particular , TLRs and RLRs may work cooperatively to induce IFN production in pDCs where the division of labor between the PRR systems may be dependent on the stimulating ligand . In order to test the ability of the pDC-GEN2 . 2 cells to control viral replication in hepatocytes , we used the JFH-1/Huh7 . 5 . 1 in vitro culture system [26] . Cell-free supernatants from pDC-GEN2 . 2 cells that had been transfected with the pU/UC RNA were added to Huh7 . 5 . 1 cells 24 hours after infection with the HCV JFH-1 virus . Four days later , Huh7 . 5 . 1 cells were lysed and the viral copy number was assayed by qRT-PCR . We found that CM from pDC-GEN2 . 2 cells transfected with the pU/UC RNA controlled viral replication better than CM from pDC-GEN2 . 2 cells transfected with the X-region RNA or mock-transfected ( Figure 5A ) . Moreover , this effect was dose-dependent . The viral control seen with the CM was attenuated at the 1∶10 dilution and lost at the 1∶100 dilution ( Figure 5B ) . In order to identify which signaling pathways were activated within hepatocytes following co-culture with CM from pDC-GEN2 . 2 cells , we used a JAK/STAT pathway gene expression array ( Table S1 ) . Huh7 . 5 . 1 cells were infected for 24 hours , and then CM was added at a 1∶1 dilution . Sixteen hours after the addition of the conditioned media , we isolated RNA and examined the gene expression using the JAK/STAT pathway gene expression array . The time-point of 16 hours was selected since the peak for Type I IFNs is 6 hours and the peak for Type III IFNs was 24 hours [27] . The most up-regulated genes were STAT1 and IRF-9 ( ISGF3G ) , a key regulator in the JAK/STAT pathway , and these were confirmed by individual qRT-PCR ( Figure S4 ) . Next , we focused on defining the relative contribution of Type III IFNs secreted by the pDC-GEN2 . 2 cells to viral control . In our model system , at the same time as the addition of the 1∶1 dilution of CM from the pDCs to the infected Huh7 . 5 . 1 cell cultures , we added a blocking antibody against Interferon-λ1/λ3 ( IL-29/28B ) or an isotype control antibody or the Vaccinia virus protein B18R . Using the Vaccinia virus protein B18R , which only blocks Type I IFNs and does not recognize IFNλ [28] , viral control induced by the CM was eliminated ( Figure 5C ) . When blocking the Type III IFNs with the dual blocking antibody , viral control was impaired . In the presence of CM and the isotype antibody , there was greater than 74% viral control , and this effect was partially lost when the dual blocking antibody was present ( Figure 5D ) . This finding suggests a role for Type III Interferons in the control of viral replication . This was further confirmed by addition of recombinant Type III IFNs in the absence of CM at levels comparable to those detected within the supernatants of HCV PAMP-transfected pDC-GEN2 . 2 cells ( Figure 5E ) ; all recombinant Type III IFNs alone demonstrated viral control at these experimentally-determined concentrations . In order to address the ability of the pDC-GEN2 . 2 cells to respond to intact virus and infected Huh7 . 5 . 1 cells we co-cultured the pDC-GEN2 . 2 cells with infected Huh7 . 5 . 1 . After 24 hours of exposure to the infected Huh7 . 5 . 1 cells , the pDC-GEN2 . 2 cells had significantly upregulated Type I and Type III IFNs compared to co-culture with uninfected Huh7 . 5 . 1 cells , represented by IFNB1 and IL-29/IFNλ1 ( Figure 6A ) . In contrast to the HCV PAMP stimulation , we did not see upregulation of RIG-I by PCR ( data not shown ) . Additionally , infected Huh7 . 5 . 1 cells were co-cultured with pDC-GEN2 . 2 cells that had been transfected with the pU/UC tract RNA and examined for HCV copy number . While the presence of pDC-GEN2 . 2 cells alone did lead to a reduction in viral copy number , co-culture with pDC-GEN2 . 2 cells transfected with the pU/UC tract RNA controlled virus significantly better than pDC-GEN2 . 2 cells that were transfected with the X-region RNA or mock-transfected ( Figure 6B ) . We wanted to confirm the effects seen in the pDC-GEN2 . 2 cell line in response to the HCV PAMP in ex vivo human pDCs . To accomplish that , we isolated pDCs from four normal healthy donors ( see Methods ) . We characterized the pDCs using flow cytometry ( Figure S5 ) and found these cells to be 95+% pure with little contamination from natural killer cells , T cells , B cells or monocytes . These cells also expressed low levels of the co-stimulatory markers CD80 and CD86 but highly expressed CD44 . After stimulation with the HCV RNA in the exact same way as the pDC-GEN2 . 2 cell line ( 1 µg of HCV PAMP RNA , either pU/UC or X-region RNA , was transfected into the cells; RNA was isolated 8 hours post transfection and assayed for expression of IFN genes ) , the ex vivo pDCs significantly upregulated IFNα2 ( except the TT subject ) , IFNβ1 , IL-28A ( IFNλ2 ) , IL-28B ( IFNλ3; except the CT subject ) and IL-29 ( IFNλ1 ) compared to the X-region RNA ( Figure 7A ) . Due to the importance of the IL-28B ( IFNλ3 ) SNP to HCV clearance , we selected subjects with each of the 3 possible genotypes to test . The subjects with the favorable CC genotype had over 100 fold more robust responses to the pU/UC RNA than the subjects with the CT or TT genotype . Additionally , the CC genotype subjects' gene expression levels were significantly higher than the non-CC genotype subjects' gene expression level in all cases except IL28B/IFNλ3 ( IFNα2 p<0 . 05 , IFNβ1 p<0 . 01 , IL28A/IFNλ2 p<0 . 01 , IL28B/IFNλ3 n . s . , IL29/IFNλ1 p<0 . 01; Figure 7B ) . While the TT genotype had the least robust response , the IFN genes were still significantly upregulated compared to the X-region RNA control . One of the reasons for the difference between the cell line Interferon levels and the non-CC ex vivo cell Interferon levels may be that the cell line has both of the “favorable” genotypes for IL-28B/IFNλ3 ( rs12979860; CC ) [14] , [15] and RIG-I ( rs10813831; AA ) [18] while the non-CC subjects have the “unfavorable” TT or CT for the IL-28B/IFNλ3 SNP and AA or GG for the RIG-I SNP . Though the AA RIG-I genotype is considered to be “favorable” , the presence of that favorable allele was not enough to overcome the “unfavorable” IL-28B/IFNλ3 SNP . These data confirm that ex vivo pDCs are capable of responding robustly to HCV RNA with Type I and III Interferon production . The supernatants from these cells were assayed for IFNα and IL-29 ( IFNλ1 ) by ELISA ( Figure S5A and S5B ) . Each subject's pDCs had more IFN present in the pU/UC-transfected condition compared to the X-region-transfected condition . Conditioned media from the transfected primary ex vivo pDCs were used to treat infected Huh7 . 5 . 1 cells in the same manner as previously described for the pDC-GEN2 . 2 CM . The pU/UC-transfected CM from the CC subjects ( but not the non-CC subjects ) controlled virus better than X-region-transfected CM from those same subjects ( Figure S5C ) .
Dendritic cells play a critical role as innate pathogen sensors [29] . Although they constitute only 0 . 2% to 0 . 8% of peripheral leukocytes in healthy subjects , pDCs have been found to produce over 95% of Type I IFNs by PBMCs in response to many viruses [29] . HCV represents the most common blood-borne viral infection for which there is no vaccine [6]; acute infection is followed by development of viral persistence in the majority of subjects . In this study , we comprehensively characterized the response to transfected hepatitis C viral PAMP using the pDC cell line GEN2 . 2 . As with any cell line , there were some differences between the cell line and ex vivo cells . In keeping with prior reports [30] , we did not find freshly isolated human ex vivo pDCs to express TLR3 ( Figure S5 ) ; in contrast , the pDC cell line did express TLR3 . However , for the most part , the cell line resembled ex vivo cells . The current dogma is that TLR and not RLR signaling predominate within the pDC population [5] . In this study , TLR ligand stimulation of pDCs increased transcription of Type I and III IFNs , as well as PRRs . These data suggest that TLR ligation not only induces IFN , but also primes the cells for recognition of additional viral stimuli . This type of priming has not been thoroughly investigated for viruses although the presence of bacterial products such as LPS have been shown to prime various innate immune cells , primarily neutrophils [31] and macrophages [32] . A feedback mechanism of TLR3 upregulation by Type I IFNs to amplify the response to TLR3 ligands was previously proposed [33] . It may also be the case that viral stimulation primes pDCs for additional recognition and enhanced IFN production . When we stimulated the cells with PEG-IFN-α2 , we failed to induce Type III IFNs , data that are consistent with a recent study using primary human hepatocytes [27] . In accordance with prior reports , we found that the cytosolic receptor RIG-I was IFN-inducible [5] . We speculate that the induction of the IFN genes may lead to an antiviral state within in vivo pDCs thus making them resistant to HCV infection and prone to rapid clearance of the virus within the pDCs themselves . This may explain why HCV RNA is not consistently found within circulating pDCs from HCV-infected patients . The pDC cell line produced IFN in response to pU/UC RNA , but not X-region RNA , with most genes peaking in expression at 8 hours , consistent with the known kinetics of IFN gene expression [34] . Given that the pU/UC RNA was generated using an in vitro transcription system , one potential concern was that the 5′ triphosphate added by the T7 polymerase might artificially induce IFN responses . However , since the X-region RNA was made in the same way ( and thus had the same triphosphate ) , but failed to induce IFN production ( Figure S2 ) , we felt confident that the gene and protein expression levels that we observed were due to pU/UC RNA-specific features . In this study , transfection was used to introduce HCV RNA into the pDC-GEN2 . 2 cells . Previous work has demonstrated that pDCs can sense HCV and respond with IFNα production in an in vitro system using Huh7 . 5 . 1 c2 cells and JFH-1 requiring cell-to-cell contact [13] . Chisari's group also showed that receptor-mediated entry of the virus into the pDCs was not required to induce IFN production [13] . This latter study provides rationale for using transfection to mimic the non-receptor mediated transfer of RNA from hepatocytes to pDCs . We found that HCV PAMP transfection of pDCs induced an antiviral effect when co-cultured with infected hepatocytes . In keeping with and extending Takahashi's study , we found that co-culture with infected Huh7 . 5 . 1 induced transcriptional upregulation of Type I and III IFNs . There are a number of means by which pDCs could sense in vivo infected hepatocytes , including encounter of viral RNA in debris from infected cells killed by immune mechanisms , such as natural killer cells , or from endocytosed virions . Transfection acts as a laboratory tool to imitate viral RNA transfer in vivo . Viral control in our replication model system related to the conditioned media ( CM ) from pU/UC RNA-transfected pDCs was partially mediated by Type III IFNs . This is in agreement with previous reports that IFNλs can control HCV in vitro [35] . We also demonstrated that STAT1 and IRF9 were upregulated by the addition of pU/UC-transfected CM , but not X-region-transfected CM , which implicate a role for IFNs in mediating the viral control within the infected hepatocytes . At the same concentrations as detected in the CM , Type III IFNs were able to control HCV replication . Both IL-28A ( IFNλ2 ) and IL-29 ( IFNλ1 ) at higher concentrations ( 100 ng/ml ) have been shown to significantly reduce HCV replication with the same efficacy and comparable to IFNα [36] . In light of prior work indicating that viral transcription is not required for RIG-I activation [37] , our collective data suggest an IFN-mediated mechanism that provides inhibition of HCV replication during the initial stages of infection [35] , [38] . Human pDC sense conserved regions of HCV RNA , resulting in the upregulation of the cytosolic RIG-I- helicase , ISGs , and robust production of Type III IFNs that mediate antiviral activity in HCV-infected hepatocytes . The current observations also help shed light on the puzzling lack of association between IL-28B/IFNλ3 SNPs and hepatic expression of Type III IFNs reported in the literature [7] . Our data using both a cell line and freshly isolated ex vivo cells suggest that pDCs sensing HCV RNA produce different levels of IFN , perhaps indicating a functional role of the polymorphisms in the IL28B/IFNλ3 gene locus . Hepatic mRNA ( in prior studies ) would have been derived predominantly from hepatocytes , whereas the main cellular sources of Type III IFNs are pDCs . Interestingly , pDCs derived from subjects with the favorable CC genotype also demonstrate significantly more robust Type I IFN transcription following HCV PAMP stimulation that may be the result of an amplification feedback loop or modification of transcriptional factors . Clearly , further work is warranted to understand the mechanisms of how Type I and III IFNs might be co-regulated . Induction of RIG-I mediated responses by the use of synthetic agonists may have implications for novel therapeutic approaches for this common infection . Moreover , mechanisms that antagonize RIG-I-mediated stimulation of IFN warrant further investigation in order to understand why the majority of subjects develop viral persistence .
Patients were recruited and written informed consent was obtained using COMIRB approved protocol # 08-0364 from the Denver Metro area . GEN2 . 2 cell line was grown on Huh7 . 5 . 1 cells in either DMEM with 10% Human Serum+1X Pen/strep+1X NEAA ( Non-essential Amino Acids ) or RPMI with 10% FBS+1X Gentamycin . No differences in the functionality of the cells were seen based on media condition ( Data not shown ) . Briefly , Huh7 . 5 . 1 cells were plated and allowed to grow until 90% confluency then the GEN2 . 2 cells were added . For experiments , the non-adherent fraction of the culture was used . These cells were over 95% BDCA-2+ and CD45+ thus indicating that the feeder line was not included in these experiments . Cells were fixed in 4% PFA for 30 minutes . Cells were then washed in DPBS and resuspended in blocking buffer ( 3% BSA , 10% FBS , 0 . 3% Triton X-100 in DPBS ) for 30 minutes at room temperature . Cells were then washed in DPBS and resuspended in Phalloidin conjugated to AlexaFluor 488 ( Invitrogen A12379 ) in Binding Buffer ( 10% FBS , 0 . 3% Triton X-100 in DPBS ) for 30 minutes in the dark at room temperature . Cells were then washed in DPBS and resuspended in DPBS . Cells were cytospun on the slide by using adding the cell suspension to a cytofunnel and spun at 1000 rpm for 5 min . The slide was allowed to dry at room temperature in the dark . Mounting media containing DAPI ( Vector Labs H-1200 ) was added to the slides and covered with a coverslips and sealed . Slides were viewed on a Leica DM5000B with a Leica DFC350FX camera . The objective lenses were Leica 506507 ( 10X/0 . 30 NA ) , Leica 506506 ( 20X/0 . 50 NA ) and Leica 506145 ( 40X/0 . 75 NA ) . Images were taken at room temperature ( 23°C ) and acquired using Leica FW400 software . Images were not adjusted or modified after image capture . Cells were plated in a 24 well low-adherence plate ( Costar 3473 ) at a concentration of 1×106 cells in 1 mL per well . TLR ligands , IFNα or media were added: either alone or in combination: Poly I:C [100 µg/mL , Sigma P0913] , ODN 2216 CpG [250 µM , Invivogen tlrl-hodna] , Loxiribine [1 mM , Invivogen tlrl-lox] , PEG-IFNα [100 ng/mL] . The cells were incubated at 37C , 5% CO2 for 6 hours at which time RNA was isolated using RNeasy Mini kit ( Qiagen , 74106 ) with the Qiashredder option ( Qiagen , 79656 ) . RNA was quantified using a Nanodrop microspectometer . 1 µg of RNA was used to make cDNA using Quantitect RT kit ( Qiagen , 205313 ) . qRT-PCR was performed using SYBR Green primers and master mix from Qiagen and run on a StepOnePlus qPCR machine from Applied Biosystems . Data was analyzed by the ΔΔCT method . All primers used in the qRT-PCR assays were purchased from Qiagen . pU/UC and X-region plasmids were kindly provided by Dr . Michael Gale's lab . The plasmids were amplified using PCR ( X-region Forward 5′-TAATACGACTCACTATAGGTGGCTCCATCTTAGCCCTA-3′ X-region Reverse 5′-ACTTGATCTGCAGAGAGGCCAGTATCA-3′ HCV pU/UC Forward 5′-TAATACGACTCACTATAGGCCATCCTGTTTTTTTCCC-3′ HCV pU/UC Reverse 5′-AAAGGAAAGAAAAGGAAAAAAAGAGG-3′ ) with a high fidelity polymerase ( Invitrogen 11304-011 ) . The PCR products were separated with electrophoresis on an ( 1% ) agarose gel . The bands of interest were extracted from the gel ( Gel extraction kit , Qiagen 28704 ) and transcribed in vitro ( Applied Biosystems AM1354M ) . The final product was quantified using a Nanodrop microspectometer . For removal of the 5′ Phosphate groups , the HCV PAMP RNA was treated with Antarctic phosphatases ( New England Biolabs , M0289 ) as per manufacturer's instructions . The enzyme was deactivated for 5 min at 65C . This RNA was then used as described for the untreated HCV PAMP ( see below ) . Cells were plated in a 24 well low-adherence plate at a concentration of 1×106 cells in 1 mL per well . 1 µg of pU/UC , or X-region RNA or 1 µL water ( mock ) were transfected ( Mirus 2250 ) for 2 , 4 , 8 or 24 hours at 37C , 5% CO2 . 1 µg of RNA was used because 1 µg of HCV PAMP RNA was used by Dr . Gale's group in their original study of the immunogenic capacity of these RNAs [16] . RNA was isolated , cDNA was generated and qRT-PCR was run and analyzed as described for TLR stimulation . Supernatants were also collected at 2 , 4 , 8 or 24 hours post-transfection . Cells were plated in a 24 well low-adherence plate in 500 µL/1×106 cells serum-free Accell Delivery Media ( Dharmacon # B-005000 ) . siRNA was then added to a final concentration of 750 nM . Cells were incubated for 48 hours at 37C , 5% CO2 . Accell Delivery Media+10% Heat-inactivated FBS ( 500 µL/1×106 cells ) was added to the wells . Cells were incubated for an additional 24 hours and then stimulated as described above . DDX58 ( RIG-I ) siRNA ( Smart Pool; E-012511-00 ) and scrambled siRNA ( Non-targeting control; D-001910-10 ) were obtained from Fisher Scientific . ELISA kits for IFNα ( PBL Interferon Source , 41105 ) , IFNβ ( PBL Interferon Source , 41410 ) , IL-28A/IFNλ2 ( RayBio , ELH-IL28A-001 ) and IL-29/IFNλ1 ( eBiosciences , 88-7296 ) were used as per the manufacturer's instructions . All samples used at either a 1∶1 or 1∶10 dilution and were incubated overnight at 4C . Cell lysates were prepared by stimulating cells as described above . Cells were harvested on ice and washed twice with cold PBS . Cells were then lysed with a Modified RIPA buffer ( Tris-HCl 50 mM , 1% NP-40 , 0 . 25% Na-deoxycholate , NaCl 150 mM , EDTA 1 mM pH 7 . 4 ) with 1X protease inhibitors ( Roche 11 836 170 001 ) and 1X phophatase inhibitors ( Fisher Scientific , PI-78420 ) . After 30 minutes of agitation at 4C , cell lysates were centrifuged and pellets were discarded . Protein levels were assayed using a BCA assay ( Fisher , PI-23221 and PI-23224 ) as per manufacturer's instructions . Samples were separated using SDS-PAGE on Mini-protean TGX Any kD gels ( Bio-rad , 456-9033 ) and transferred onto a nitrocellulose membrane using a wet transfer system . Membranes were blocked for 1 hour with 5% milk in TBST or PBST , washed , and proteins were analyzed by immunoblotting with standard methods using antibodies specific to RIG-I ( Enzo Lifesciences ) , MDA5 ( Cell Signaling Technology ) , MyD88 ( Cell Signaling Technology ) , IL-28B/IFNλ3 ( R&D Systems ) , RNase L ( Thermo Scientific ) , TRIF ( Cell Signaling Technology ) , tubulin ( Sigma ) , and GAPDH ( Abcam ) . Secondary antibodies conjugated to HRP were obtained from Jackson ImmunoResearch and Cell Signaling Technology , and immunoreactive bands were detected with either the Immuno-Star HRP Substrate kit ( Bio-Rad , 170-5040 ) or the Amersham ECL Prime Western Blotting Detection Reagent ( GE Healthcare ) . Densitometry was performed using ImageJ ( NIH ) , and proteins of interest were normalized to a reference protein ( Tubulin or GAPDH ) . Whole RNA was isolated from HCV PAMP stimulated cells as described above . RNA was run on a 1% agarose formaldehyde gel with Ethidium bromide . Gel was visualized via UV . Huh7 . 5 . 1 cells ( kindly provided by Frank Chisari , Scripps ) were plated in a 24 well plate at a concentration of . 125×106 cells per well in 1 mL media overnight . The cells were then infected with JFH-1 at an MOI ( multiplicity of infection ) of 0 . 01 . Twenty-four hours following infection , the supernatants generated from the HCV PAMP transfection ( either Mock ( negative ) transfection , X-region or pU/UC ) were added at various dilutions . Five days post-infection the supernatants were aspirated and RNA was isolated as previously described . HCV copy number qRT-PCR was performed using JFH-1 primers ( JFH-1 Forward 5′-CGACACTCCGCCATGAATCACT-3′ and JFH-1 Reverse 5′-CACTCGCAAGCGCCCTATCA-3′ ) and TaqMan TAMRA probe ( 5′-6FAMAGGCCTTTCGCAACCCAACGCTACTTAMRA-3′ ) . The copy number was determined using a standard curve . Using the infection control condition absolute HCV copy number all other conditions' copy numbers were normalized for that particular experiment . The following formula was used to determine Normalized HCV Copy Number: Normalized HCV Copy Number = ( Absolute copy number for condition/absolute copy number for infection control ) . A blocking antibody against IL-28B/IL-29 ( IFNλ3/IFNλ1 ) or an isotype control antibody were obtained from R&D Systems ( 10 µg/mL , MAB15981 , MAB003 ) or the recombinant Vaccinia protein B18R from eBioscience ( 0 . 1 µg/mL; 14-8185-62 ) and added to the supernatants at 24 hours post-infection and the experiment was performed as described . Recombinant Interferon proteins ( IL-28A/IFNλ2 , PeproTech , 300-02K: 1500 pg/mL; IL-29/IFNλ1 PeproTech , 300-02L: 500 pg/mL; IL-28B/IFNλ3 , R&D Systems , 5259-IL-025: 10 pg/mL ) were added to the infected Huh7 . 5 . 1 cultures 24 hours post-infection and the experiment was performed as described above . For the JAK/STAT pathway gene expression arrays ( Applied Biosystems , 4414156 ) and validation Huh7 . 5 . 1 cells were plated , infected and treated with pDC supernatants as described above . Sixteen hours after the addition of the supernatants , the cells were harvested for RNA as previously described . cDNA was made and the gene expression array plates were run as per manufacturer's instructions or individual genes were examined by qRT-PCR as described above . For co-culture experiments , Huh7 . 5 . 1 cells were plated overnight and infected as described above . The non-transfected pDC-GEN2 . 2 cells were added 24 hours post-infection . Cells were co-cultured for 24 hours and then the non-adherent fraction was collected for qRT-PCR analysis . This fraction was examined for BDCA-2 expression and 98+% were BDCA-2+ indicating little to no contamination of the Huh7 . 5 . 1 cells . For the co-culture experiments with the transfected pDC-GEN2 . 2 , Huh7 . 5 . 1 cells were plated and infected as described above . Twenty-four hours post-infection , pDC-GEN2 . 2 cells were transfected with HCV PAMP RNA as described previously in the methods ( 8 hour transfection ) and added to infected Huh7 . 5 . 1 cells . As described above for pure Huh7 . 5 . 1 cultures , the cultures were examined five days post-infection for HCV copy number by qRT-PCR . Patients were recruited and consented using COMIRB approved protocol # 08-0364 from the Denver Metro area . PBMCs from normal , healthy patients were screened for pDC percentage of >0 . 25% of total PBMCs using Lineagenegative , HLA-DR+ , BDCA-2+ and BDCA-4+ . Patient cells were also genotyped for IL28B and RIG-I SNPs ( rs12979860 and rs10813831 , respectively ) . Patients were leukopheresed and pDCs were immediately isolated using Miltenyi's Diamond Plasmacytoid Dendritic Cell Isolation Kit II ( Miltenyi Biotech , 130-097-240 ) and characterized using flow cytometry and fluorescent microscopy as described for the pDC cell line . Cells were then stimulated in the exact same way as the pDC-GEN2 . 2 cell line with the HCV PAMP RNA ( 1 µg of RNA was transfected into the cells , 8 hours later RNA was isolated , 1 µg of RNA was used to make cDNA and assayed for gene expression by qRT-PCR ) . Due to cell number limitations , only a single time-point could be examined . Eight hours was chosen since the pDC-GEN2 . 2 gene expression peaked at this time-point . ELISAs and viral control were performed as described above for the pDC-GEN2 . 2 cell line . Statistics were performed using Graphpad Prism statistical package . Student's T-test was used for comparisons amongst groups with more than 30 data points . Mann-Whitney non-parametric test was used for comparisons amongst groups with less than 30 data points . One sample t-tests or Mann-Whitney tests were used to compare fold increases of stimulated conditions with control conditions . | Hepatitis C Virus ( HCV ) is the most common bloodborne pathogen for which no vaccine is available . Infection with the virus often leads to persistent ( or chronic ) infection . Patients with chronic HCV infection can develop progressive liver disease and liver failure , leading to the need for a transplant . It is not fully understood why some people clear the virus and others develop persistent infection . Understanding differences in how patients respond to the virus in the early phases of infection may lead to better treatment of HCV . Here , we use a highly conserved region of the HCV genome to examine innate immunological responses to HCV . We found that plasmacytoid dendritic cells , innate cells keyed to respond with anti-viral interferon proteins , recognize the virus . Additionally , we show that pDCs use RIG-I in the recognition of this virus , which was previously thought to be dispensable in pDCs . The proteins secreted by these cells can control viral replication in a cell-based laboratory system . In cells isolated from healthy donors , we found that fresh human cells can respond in the same manner to the virus as the laboratory strain of cells , and there was a correlation with genetic differences . Our study offers novel insight to how the body recognizes HCV during early infection and host-virus interactions that mediate viral control of this common infection . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"hepatitis",
"c",
"medicine",
"infectious",
"hepatitis",
"immunity",
"gastroenterology",
"and",
"hepatology",
"innate",
"immunity",
"immunity",
"to",
"infections",
"immunology",
"biology",
"liver",
"diseases"
] | 2013 | Hepatitis C Virus Pathogen Associated Molecular Pattern (PAMP) Triggers Production of Lambda-Interferons by Human Plasmacytoid Dendritic Cells |
Polycomb Repressive Complex ( PRC ) 1 and PRC2 regulate genes involved in differentiation and development . However , the mechanism for how PRC1 and PRC2 are recruited to genes in mammalian cells is unclear . Here we present evidence for an interaction between the transcription factor REST , PRC1 , and PRC2 and show that RNF2 and REST co-regulate a number of neuronal genes in human teratocarcinoma cells ( NT2-D1 ) . Using NT2-D1 cells as a model of neuronal differentiation , we furthermore showed that retinoic-acid stimulation led to displacement of PRC1 at REST binding sites , reduced H3K27Me3 , and increased gene expression . Genome-wide analysis of Polycomb binding in Rest−/− and Eed−/− mouse embryonic stem ( mES ) cells showed that Rest was required for PRC1 recruitment to a subset of Polycomb regulated neuronal genes . Furthermore , we found that PRC1 can be recruited to Rest binding sites independently of CpG islands and the H3K27Me3 mark . Surprisingly , PRC2 was frequently increased around Rest binding sites located in CpG-rich regions in the Rest−/− mES cells , indicating a more complex interplay where Rest also can limit PRC2 recruitment . Therefore , we propose that Rest has context-dependent functions for PRC1- and PRC2- recruitment , which allows this transcription factor to act both as a recruiter of Polycomb as well as a limiting factor for PRC2 recruitment at CpG islands .
Polycomb group ( PcG ) proteins are epigenetic regulators of gene expression and play an essential role during embryonic development [1] . The Polycomb repressive complex 2 ( PRC2 ) is the only known enzyme that mediates di- and tri-methylation of histone H3 on lysine 27 ( H3K27Me2/3 ) , modifications believed to be required for PcG-mediated gene repression [2] , [3] , [4] , [5] . PRC2 consist of three core components , Ezh2 , Suz12 and Eed , which are all required for early mouse development [6] , [7] , [8] . H3K27Me3 can function as an epigenetic mark for the recruitment of PRC1 , a large heterogenous complex [9] , which among others include the Cbx- and Rnf2 ( Ring1B ) proteins . Rnf2 catalyzes the ubiquitination of histone H2A on lysine 119 ( H2AK119Ubi ) [10] , [11] and as for the members of the PRC2 complex , disruption of the Rnf2 gene in mouse causes a similar developmental phenotype with arrest at gastrulation [12] . Furthermore , Rnf2 has recently been shown to be part of at least two additional gene regulatory complexes , the E2F6 . com-1 complex [13] and the Fbxl10-BcoR complex [14] . The importance of PcG protein complexes in stem cell maintenance and differentiation has been extensively studied in mouse embryonic stem ( mES ) cells . Previous work have shown that genetic elimination of either PRC1 or PRC2 function , by knockout of Rnf2 or Eed [12] , [15] , [16] , [17] , [18] leads to derepression of several lineage-specific genes , that tend to destabilize mES cells , although they still preserve their ability to self-renew and differentiate . Interestingly , the Wutz laboratory showed , that the simultanous loss of both the PRC1 and PRC2 complexes in mES cells abrogates differentiation [19] . In the same study it was suggested that the PRC1- and PRC2-complexes can function independently to repress a common set of genes important for stem cell maintenance . There are several models explaining how the PcG complexes are recruited to specific target genes . In mammalian cells , it is a long-standing dogma that the PRC2 complex is recruited to target gene promoters by so far unidentified transcription factors ( TFs ) , through the recognition of the underlying DNA sequences . Subsequently , this leads to H3K27 methylation and recruitment of PRC1 through the chromodomain-containing CBX proteins [20] , [21] . Several recent publications suggest that Jarid2 can function as a PRC2 recruitment factor [22] , [23] , [24] , [25] , [26] . Furthermore , non-coding RNAs ( ncRNA ) interact with both PRC1 and PRC2 and seem to be working in parallel or in combination with TFs in the ability to recruit the PcG complexes to genomic loci [27] , [28] , [29] . Interestingly , recent data have implicated ncRNAs as being important for the recruitment of PRC2 to CpG islands and suggested that these genomic entities are sufficient for PRC2 recruitment [30] , [31] . In Drosophila , several TFs are involved in recruiting the PcG complexes to DNA elements called Polycomb Response Elements ( PREs ) [32] , [33] , [34] . However , among these transcription factors only an ortholog of Pho , YY1 , is preserved in mammals and has been found to interact with PRC1- [35] and PRC2-subunits [36] . Based on the data from Drosophila , it is therefore likely that several TFs beside YY1 are involved in the recruitment of PcG complexes in mammalian cells and identification of such factors is needed in order to define mammalian PREs . We now present evidence for an interaction between the TF REST ( also called Neuron-Restrictive Silencing Factor , NRSF ) and the PRC1- and PRC2-complexes , which is independent of RNA . By shRNA mediated knockdown , DNA microarray expression analysis and ChIP analysis , we show that a number of genes are co-regulated by REST and RNF2 in human teratocarcinoma NT2-D1 cells . Genome-wide analysis by ChIP-sequencing ( ChIP-seq ) in mES cells reveal the co-existence of PRC1 and PRC2 on Rest binding sites . Combining our biochemical interaction analysis of Rest-PcG complexes and the genome-wide analysis of Rest-PcG binding sites , suggest that Rest is required for the recruitment of PRC1 and PRC2 to a subset of its target genes in mES cells . Interestingly , the recruitment of Rnf2 to Rest binding sites can occur independently of both CpG islands and PRC2 activity . Surprisingly , PRC2 was frequently increased around Rest binding sites located in CpG rich regions in the Rest−/− mES cells , which suggest that other Rest-associated activities can limit PRC2 recruitment . Based on these observations we propose that Rest has context-dependent functions for PRC1- and PRC2-recruitment to target genes in mammalian cells and that PRC1 is a co-repressor for Rest .
We were interested to examine whether the transcription factor REST and the PRC1 complex would interact in vivo , encouraged by previous observations , where we identified REST in a double-tag purification of the CBX8 interacting protein , HAN11 ( WDR68 ) ( Figure S1; see the information in Procedure S1 ) . We performed size-exclusion chromatography of nuclear extracts from the human teratocarcinoma cells , NT2-D1 ( Figure 1A , left part ) and HEK 293 cells ( Figure S1E ) and performed immunoprecipitation of REST from pools of different fractions as indicated ( Figure 1A , right part ) . The data showed , that the PRC1 core subunits , RNF2 , BMI1 , NSPC1 and CBX8 co-immunoprecipitated with REST from high-molecular weight fractions ( F7–9 , labelled pooled fractions 1 ) . In addition to CBX8 , we also detected CBX7 when performing the REST immunoprecipitation on the pool of fractions F11–13 ( labelled pooled fractions 2 ) . Importantly , we did not detect E2F6 , BCOR or HP1γ in the REST immunoprecipitations from NT2-D1 cells , showing that the complex that we found associated to REST , contains members of the canonical PRC1 complex and differs from the previously described E2F6 . com-1 [13] and Fbxl10-BcoR complexes [14] , [37] . For simplicity we will refer to this REST associated PcG complex as PRC1 throughout the remaining part of this work . Importantly , beside core components of the PRC1 complex , we furthermore found that the PRC2 subunits EZH2 and SUZ12 co-immunoprecipitated with REST from pool 1 and 2 . Since it has recently been shown that the long non-coding RNA HOTAIR can recruit CoREST/REST/LSD1- and PRC2 complexes through the 5′ and 3′ends respectively [27] , we checked whether RNase treatment of our immunoprecipitates would dissociate PcG complexes from REST . As seen in Figure 1B , degradation of single- and double-stranded RNA had no effect on the interactions between endogenous REST , PRC1 and PRC2 . Furthermore , we found that the interactions between REST and the Polycomb complexes were not due to DNA bridging , as these were not eliminated by ethidium bromide treatment of immunoprecipitated complexes ( Figure S1D ) . To confirm the interaction between Rest and the PRC1 and PRC2 complexes in mouse embryonic stem ( mES ) cells , we performed immunoprecipitations of Rest on pooled fractions obtained from size-exclusion chromatography of nuclear extracts from Wt , Eed−/− and Rnf2−/− mES cells ( Figure 1C and 1D ) . Similarly to the results obtained in NT2-D1 cells , we found that Rest interacts with core subunits of the PRC1 and the PRC2 complexes in mES cells . When comparing the relative efficiency of co-immunoprecipitation in Wt and Eed−/− mES cells , we furthermore found that the interaction between Rest and the PRC2 subunits , Ezh2 and Suz12 , did not require Eed . As expected , the total level of the other two core subunits of PRC2 , Suz12 and Ezh2 , were somewhat reduced in the Eed−/− cells ( due to destabilization ) , which was reflected in diminished amounts of these two subunits in the Rest IPs from the Eed−/− mES cells . Furthermore , there was a clear change in size distribution of the PRC2 complex in the Eed−/− mES compared to the Wt cells , showing reduced amounts of the most high-molecular weight forms ( Pooled fractions 1 , Figure 1C , upper panels ) . This was reflected in the co-immunoprecipitation of Ezh2 and Suz12 by Rest , where the binding of these two subunits to Rest , in the absence of Eed , was evident in the IP from pool 2 , but very reduced in the IP from pool 1 ( Figure 1C , lower panel ) . Moreover , the absence of Eed did not affect the efficiency of co-immunoprecipitation between Rest and the PRC1 subunits , Rnf2 and Nspc1 ( Figure 1C ) . Interestingly , the Nspc1 subunit was more abundant in the Eed−/− mES cells ( see input lysates ) , which was also reflected in the Rest IPs when comparing Wt and Eed−/− mES cells ( Figure 1C , lower panel ) . Immunoprecipitations performed on fractions from Rnf2−/− mES cells furthermore revealed that Rnf2 is not required for the interaction between Rest and the PRC1 subunit Nspc1 . This suggested that the interaction between Rest and PRC1 was not mediated directly through Rnf2 . In conclusion , our biochemical data clearly showed that REST interacts with PRC1 and PRC2 protein complexes in mammalian cells and that these REST-PcG complexes are independent of non-coding RNAs . The interaction between REST and PRC1 suggested that the PRC1 complex might function as a co-repressor for REST . To investigate this , we performed a DNA microarray expression analysis to compare the effect of shRNA mediated knockdown of REST with the knockdown of RNF2 ( Figure 2A and 2B ) . We decided to use the human NT2-D1 cells , since these can easily be induced to differentiate along the neuronal path by retinoic acid ( RA ) stimulation [38] . As summarized in the cluster analysis and the Venn diagram , a highly significant number of genes were co-regulated by REST and RNF2 ( Figure 2B ) ( see the information on data handling in Procedure S1 ) . While REST knockdown lead to significantly increased expression of 1 , 862 genes ( >2 fold ) , RNF2 knockdown lead to increase in the expression of 775 genes ( >2 fold ) ( Figure 2B and Table S1 ) . Among these genes , 258 were found to be co-regulated , corresponding to more than 30% of all the genes up-regulated in response to RNF2 knockdown . Although , the majority of the co-regulated genes were up-regulated , we also found that a considerable number of genes were co-down-regulated , suggesting that we not only observe direct effects , but also indirect effects on expression . In agreement with previous reports on REST- and PcG targets , co-up-regulated genes enrich for developmental functions , while co-down-regulated genes showed no significant enrichment for any particular biological function ( Table S9 ) . This suggests that among the co-up-regulated genes , there was enrichment for genes that were directly targeted by the REST-PRC1 complex . To compare the influence of RNF2 to a well-described REST-interacting co-repressor complex , the CoREST complex [39] , we decided to perform a new shRNA mediated knockdown experiment using constructs for REST , RNF2 and RCOR1 ( gene coding for the CoREST protein ) ( Figure 2C and 2D ) . Using QPCR , we analyzed the change in expression of 7 different genes from the group of 258 co-regulated genes identified in the microarray ( Figure 2B ) . Interestingly , the data suggest that the genes can be divided into two groups . The first group contains genes ( MEIS1 , HOXD11 , DLX5 and TBX3 ) that were co-regulated by REST and RNF2 but only slightly affected by RCOR1 knockdown ( Figure 2C ) . The second group includes genes ( GAD1 , MEIS2 and HES1 ) that were co-regulated by REST and RNF2 , but furthermore showed an increase in expression upon RCOR1 knockdown ( Figure 2D ) . We also found an example of a gene ( CALB1 ) that was co-regulated by REST and CoREST , but unaffected by RNF2 knockdown ( Figure 2D ) . To confirm the interaction between REST and RNF2 on co-regulated target genes we performed ChIP for two of the genes in the first group ( MEIS1 and HOXD11 ) and one gene in the second group ( MEIS2 ) in NT2-D1 cells treated with shRNAs for REST , RNF2 or empty control shRNA ( Figure 2E ) . As expected , REST knockdown reduced the level of REST at all three target genes . Importantly , REST knockdown also lead to a reduced amount of RNF2 at these genes ( Figure 2E ) . In contrast , while a shRNA against RNF2 almost completely eliminated RNF2 binding to the MEIS1 and MEIS2 genes , it had no effect on REST binding , which is in agreement with REST being a DNA binding factor recruiting RNF2 . At the HOXD11 locus there was a small , but significant , reduction of REST binding in response to RNF2 knockdown . Interestingly , H3K27Me3 was significantly reduced in both REST- and RNF2 knockdowns , indicating that the PRC2 complex interacted with both REST and RNF2 containing complexes on the MEIS1 , MEIS2 and HOXD11 genes . Altogether we found that REST was required for the recruitment of RNF2 and the maintenance of H3K27Me3 on a selected number of neuronal genes and that REST and RNF2 co-regulate a highly significant number of genes in NT2-D1 cells . To test if Rest is required for the recruitment of PRC1 and PRC2 to their target genes in mouse embryonic stem ( mES ) cells , we performed our analyses in the previously published Wt and Rest−/− mES cells [40] , [41] . We observed that the overall levels of Suz12 , Ezh2 , Rnf2 and Oct4 proteins were unchanged in the Rest−/− mES cells as compared to the Wt mES cells ( Figure 3A ) . Although , the global level of the H3K27Me3 was unchanged between the Wt and Rest−/− mES cells ( Figure 3A ) , ChIP-sequencing ( ChIP-seq ) ( see information on data handling in Procedure S1 ) for H3K27Me3 , Suz12 and Rnf2 revealed clear local differences for these factors . This is illustrated in Figure 3B by ChIP-seq profiles for three genes ( Brunol6 , Best2 and Prrxl1 ) showing co-localization between Rest , Rnf2 and Suz12 in the Wt mES cells . Notably , for two of the genes ( Brunol6 and Best2 ) there was an almost complete loss of Rnf2 binding at the center of the Rest peaks in the Rest−/− mES cells . For the Prrxl1 gene , there was a surprising increase in Rnf2 binding , suggesting that Rest can also counteract PRC1 recruitment at certain loci . Moreover , the increase in Rnf2 was accompanied by an increase in PRC1 subunit Cbx7 on the Prrxl1 gene ( Figure S2C ) . The effects on Suz12 binding were similar , but less pronounced , with reduced binding at Brunol6 and Best2 and somewhat increased binding at the Prrxl1 gene . For the distribution of H3K27Me3 , we observed more widespread effects and the signals were reduced throughout the gene body at Brunol6 and Best2 , while at Prrxl1 the signal was essentially unchanged . Moreover , the non-Rest target gene , Gjb2 , did not show changes in Rnf2 and Suz12 binding as well as H3K27Me3 levels in the Rest−/− mES cells ( Figure 3B ) . Importantly , the mRNA levels for Brunol6 and Best2 were up-regulated in Rest−/− mES cells while unchanged for Prrxl1 ( Figure 3D ) , showing a functional link between the loss of PcG proteins , the reduced H3K27Me3 mark and gene activity in the absence of Rest . We confirmed the effects observed in the ChIP-seq profiles by direct ChIP ( Figure 3C ) using primers at the Rest peak position ( as indicated by arrow heads in Figure 3B ) . To obtain a global picture of how important Rest is for the recruitment of PRC1 and PRC2 in mES cells , we plotted the average distributions of Rnf2 ( PRC1 ) and Suz12 ( PRC2 ) signals in Wt and Rest−/− mES cells from our ChIP seq analysis relative to 3 , 378 identified Rest binding sites in the Wt cells ( Figure 4A and Table S2 ) . We included Jarid2 in this analysis , since this protein has recently been shown to be important for PRC2 recruitment in mES cells [24] , [25] , [22] , [23] , [26] . These results showed that in the Wt mES cells Rnf2 , Suz12 and Jarid2 all enrich at Rest binding sites . Interestingly , Rnf2 and Jarid2 displayed focused localization at the Rest binding sites , whereas Suz12 had a broader distribution . When comparing Rest−/− to Wt mES cells we observed that the overall Rnf2 signal was reduced at the center of the Rest peaks , but was slightly increased in the flanking regions ( Figure 4A ) . Surprisingly , the overall signal for Suz12 was clearly increased in the Rest−/− mES cells , suggesting that Rest can function to limit PRC2 binding in certain contexts . Moreover , when looking at these average distributions , we observed a similar increase in the Jarid2 signal flanking the Rest binding sites in the Rest−/− mES cells ( Figure 4A ) , which is in agreement with Jarid2 being part of the PRC2 complex . As we observed reduced Rnf2 and Suz12 levels at specific Rest binding sites in the Rest−/− mES cells ( Figure 3B and 3C ) , we wanted to investigate how frequently Rest target loci had lost or gained PcG-binding . For this we generated heat-maps based on ChIP-seq tracks for Rnf2 , Suz12 and Jarid2 around individual Rest binding sites scored positive for PcG proteins either in Wt or Rest−/− mES cells ( Figure 4B and Table S3 ) . These heat-maps showed that Rnf2 , Suz12 and Jarid2 all displayed a focused signal centered at a large fraction of the Rest binding sites in Wt mES cells , whereas this focused signal was absent in the Rest−/− mES cells . To test if the PcG binding was specific for Rest binding sites , we generated matched control regions for all 3 , 378 Rest binding sites identified in Wt mES cells . Matched control regions were randomly chosen , but matched to the distribution of the Rest binding sites relative to TSS [42] . In agreement with the conclusion that Rnf2 , Suz12 and Jarid2 enrich at Rest binding sites , we found roughly four to five times as many Rest binding sites scored positive for PcG proteins compared to the matched control regions ( Figure 4D and 4E ) . Moreover , at the 395 Rest binding sites shown for Rnf2 , the signal was reduced 1 . 5 fold or more at 172 sites ( Ratio-image: blue ) and increased 1 . 5 fold or more at 45 sites ( Ratio-image: red ) in the Rest−/− mES cells compared to the Wt mES cells ( Figure 4B; upper part ) . Examples of ChIP-seq tracks for target genes from upper and lower parts of the heat-maps are shown in Figure 4C ( numbers 1–6 refers to the position in the individual heat-maps ) . For Suz12 , we found that out of 292 Rest binding sites 45 and 105 sites had a signal that was reduced or increased 1 . 5 fold or more , respectively ( Figure 4B; middle part ) . For Jarid2 , we found that out of 300 Rest binding sites , the signal was reduced 1 . 5 fold or more at 85 sites and increased 1 . 5 fold or more at 152 sites ( Figure 4B; lower part ) . Interestingly , heat-maps furthermore showed that the loss of binding in the Rest−/− mES cells , for all three proteins ( Rnf2 , Suz12 and Jarid2 ) was focused at the Rest binding site . On the other hand , increased binding was more widely distributed and not confined to the actual Rest binding site ( Figure 4B ) . This explains why we observed increased signals for PcG proteins in the regions flanking the Rest binding sites in Figure 4A . When the changes in PcG binding observed at Rest binding sites were compared to the changes at the matched control regions , we found that both the loss and the gain of signal observed at Rnf2- and Jarid2-positive Rest binding sites , as well as the gain of signal at Suz12-positive Rest binding sites , occurred much more frequently than would be expected from Rnf2- , Jarid2- , and Suz12-positive matched control regions ( Figure 4B and 4E ) . In conclusion , we found that Rest was required for the specific recruitment of Rnf2 to a considerable number of Rest binding sites , whereas Suz12 and Jarid2 showed dependency on a smaller number of loci . In addition , many loci in the Rest−/− mES cells , which bound Rest in the Wt mES cells , had increased levels of Suz12 and Jarid2 . This was also observed for Rnf2 , but less frequently , suggesting that Rest can also limit PcG protein recruitment and that other factors control how Rest affects the level of PRC1 and PRC2 at specific genomic loci . Given the biochemical interaction between Rest and Rnf2 and the enrichment of Rnf2 at Rest binding sites , it was surprising to see that Rest binding sites in Rest−/− mES cells , not always lost Rnf2 binding , but that some Rest binding sites even gained Rnf2 signal . To study this enigma further , we correlated the changes in Rnf2 binding between Wt and Rest−/− mES cells to several different parameters , such as PRC2 levels and distance to the nearest CpG island ( Figure 5A ) . This analysis showed a strong correlation between changes in Rnf2 binding and the levels of Suz12 at Rest binding sites in both Wt and Rest−/− mES cells ( Figure 5A ) . Changes in Rnf2 binding also correlated well with the distance to TSS and CpG islands , but not to the overall composition ( % CpG or %GC ) and size of the nearest CpG island ( Figure 5A ) . To analyze the PRC2 dependency for PRC1 recruitment further , we separated Rnf2-positive Rest binding sites into two groups , which either had relatively high or low Suz12 or H3K27Me3 signals in the Wt mES cells . The signal for Rnf2 in Wt versus Rest−/− mES cells was plotted in a scatter diagram ( Figure 5B ) . The results showed that Rnf2 binding was more likely to be Rest-dependent at sites with low Suz12 and low H3K27Me3 levels ( Figure 5B; blue color ) compared to sites with high signals for Suz12 and H3K27Me3 in Wt mES cells ( Figure 5B; orange color ) . In contrast , sites with high levels of Suz12 and H3K27Me3 were more likely to maintain or even gain Rnf2 binding in the absence of Rest . Furthermore , we observed that sites with increased PRC2 binding in general had increased Rnf2 levels in Rest−/− mES cells ( Figure S3 ) . To study the effect of CpG islands on changes in Rnf2 binding in the absence of Rest , we separated a subset of Rest binding sites with the strongest Rnf2 binding ( Wt mES ) into two categories: 1 ) sites with an annotated CpG island within 1 Kb ( CpG island-positive ) and 2 ) sites more than 1 Kb away from any annotated CpG island ( CpG island-negative ) ( UCSC browser ) . Using this approach we found that out of 165 Rest binding sites overlapping with Rnf2 sites app . 50% are within 1 Kb of a CpG island ( Figure 5C , 5D and Table S4 ) . Interestingly , when we plotted the average distribution of Rnf2 at CpG island-negative Rest binding sites ( >1 Kb from a CpG island ) , we observed that the Rnf2 enrichment was almost completely lost at Rest binding sites in the Rest−/− mES cells . In contrast , at CpG islands-positive Rest binding sites ( <1 Kb from a CpG island ) , we observed only a minor decrease in Rnf2 binding ( Figure 5C and 5D ) . For Suz12 , the data showed that the levels of Suz12 were clearly higher at CpG islands-positive Rest binding sites , and that the absence of Rest resulted in a marked increase in Suz12 at these sites ( Figure 5C and 5D ) . We found similar effects for Jarid2 ( Figure 5C and 5D ) , suggesting that CpG islands could be part of a mechanism responsible for the increased recruitment of PRC2 in the absence of the transcription factor Rest in the mES Rest−/− mES cells . Finally , we wanted to distinguish , whether the effect of CpG islands on Rnf2 binding was due to the presence of PRC2 at the CpG islands , or if CpG islands represented an independent entity responsible for Rnf2 recruitment . Therefore , we plotted the Rnf2Rest−/−/Rnf2Wt ratio of Rnf2-positive Rest binding sites relative to the position of the nearest TSS or the nearest CpG island , as well as to the level of Suz12 in either Wt or Rest−/− mES cells . As previously described , we observed that a larger fraction of Rnf2 positive Rest binding sites had increased Rnf2Rest−/−/Rnf2Wt ratio at positions close to TSS and CpG islands ( Figure 5E , red coloring ) , compared to those that were located far from TSS or CpG islands ( Figure 5E , blue coloring ) . Importantly , the key determinant for the change in Rnf2Rest−/−/Rnf2Wt ratios appeared to be the level of the PRC2 subunit Suz12 . Rest binding sites with high Suz12 levels and located at or close to CpG islands , had significantly increased Rnf2Rest−/−/Rnf2Wt ratios , whereas those with low Suz12 levels had a lower Rnf2Rest−/−/Rnf2Wt ratio ( Figure 5E ) . Also , a minor group of Rnf2-positive Rest binding sites , far from annotated CpG islands , had a high level of Suz12 and these had a correspondently increased Rnf2Rest−/−/Rnf2Wt ratio compared to sites with similar position and lower Suz12 levels . In summary , it appears from this analysis that 1 ) Rest was the dominant recruitment factor for PRC1 at Rest binding sites with low PRC2 levels , and 2 ) that the increased recruitment of Rnf2 observed at CpG islands in Rest−/− mES cells was an indirect effect of increased PRC2 recruitment to CpG islands and TSS-proximal Rest binding sites in the Rest−/− mES cells . The general view that PRC1 is recruited to chromatin exclusively by PRC2 activity [43] has recently been challenged , by the observation that mES cells lacking PRC2 activity still maintained H2AUbi marked histones catalyzed by the PRC1 complex [19] . Since we observed that Rest was needed for PRC1 recruitment at Rest binding sites with low levels of Suz12 and H3K27Me3 ( Figure 5B ) , we wanted to test if Rest could recruit PRC1 to Rest binding sites in cells deficient for PRC2 activity . We performed ChIP-seq for Rest and Rnf2 in E14 Wt control and Eed−/− mES cells and found that app . 90% of the Rnf2 binding sites identified in the Wt cells were lost in Eed−/− cells ( Table S6 ) . In line with PRC1 being recruited to CpG islands via PRC2 , we observed that the loss of Rnf2 binding sites were most pronounced for peaks positioned at or close to CpG islands ( Figure 6A , upper part ) , while the majority of binding sites positioned distant from CpG islands were in general maintained in the Eed−/− mES cells ( Figure 6A , lower part ) . Nonetheless , the average Rnf2 signal at Rest binding sites was only marginally reduced ( Figure 6B , upper part ) when analyzing all Rest binding sites both distant from CpG islands and at CpG island proximal regions . Notably in the Wt mES cells , we observed increased Rnf2 binding from the Rest binding sites towards the TSS , which was virtually lost in the absence of Eed ( Figure 6B , upper part ) . Furthermore , when comparing the distribution of Rest peaks in Wt versus Eed−/− mES cells , we only observed a minor effect indicating , that the overall binding of Rest to DNA was independent of PRC2 ( Figure 6B , lower part ) . In order to visualize the data summarized by the average distributions in Figure 6B , for individual genes , we generated heat-maps based on ChIP-seq tracks for individual Rest binding sites within 10 Kb of a TSS ( Figure 6C ) . In agreement with the data in Figure 6B , the wide-spread Rnf2 signals in the Wt was virtually lost in the Eed−/− mES cells , but the signal was clearly maintained at the Rest binding sites ( Figure 6C ) . To confirm the results obtained by ChIP-seq , we performed direct ChIP on four genes . We compared the Rnf2 binding at Rest binding sites in Wt and Eed−/− cells and also included a ChIP for the H3K27Me3 mark . As shown in Figure 6D , the four genes behaved differently . Gjb2 , a PcG target gene [44] showed very low levels of Rest binding in the Wt mES cells , which was not affected in the Eed−/− mES cells , while H3K27Me3 was absent and Rnf2 strongly reduced . For the Calb1 gene , a classical Rest target [45] , we confirmed a strong Rest enrichment , which correlated with enrichment for Rnf2 in the Wt mES cells ( Figure 6D ) . Even though H3K27Me3 was absent at the Calb1 locus in the Eed−/− mES cells the binding of Rnf2 was preserved ( Figure 6D ) , demonstrating PRC2- and H3K27Me3-independent recruitment of Rest-PRC1 . Nudt9 and Gpc2 , two other Rest targets in mES cells showed either a small decrease or increase , respectively , in Rest binding in the Eed−/− mES cells as compared to the Wt mES cells , while Rnf2 was not reduced . Similarly , we detected Cbx7 binding on two non-Rest PcG targets ( Gjb2 and Neurog1 ) in Wt mES cells that was lost in Eed−/− mES cells , while Cbx7 was still bound to the Stag3 gene in Eed−/− mES cells ( Figure S2 ) . In Figure 6E we present ChIP-seq profiles for the Nudt9 , Gpc2/Stag3 , and Gjb2 loci , which reflects the results obtained by direct ChIP ( ChIP followed by QPCR analysis; arrow heads indicate the position of primers used in Figure 6D ) . It is noteworthy that not all Rest binding sites were preserved between the two Wt mES cell lines ( compare Figure 3B and Figure 6E to Figure S4 ) . Since both Rnf2 and Rest binding sites in E14 Wt mES cells frequently appeared in close proximity of the TSS of genes , we asked whether Rest and Rnf2 co-localized more frequently than expected by chance in Wt and Eed−/− mES cells . We calculated the frequencies of Rest binding sites alone , Rnf2 sites alone or in combination with Rest , 20 Kb up- and down-stream of all annotated TSS positions and transcription ends ( TE ) in the mouse genome ( mm9 ) . We plotted ratios between the observed frequencies versus the frequencies expected by chance in a co-occurrence matrix ( Figure 6F ) . The observed co-occurences of Rest and Rnf2 at the TSS in Wt mES cells were actually significantly below what could be expected by chance ( red ) , whereas upstream and downstream of TSS and at TE , we observed significantly more co-occurrence than expected by chance ( blue ) ( Figure 6F ) . Remarkably and in agreement with Rnf2 binding being preserved at Rest binding sites in the Eed−/− mES cells , we observed much higher co-occurrence than expected by chance both up- and down-stream of the TSS and TE in the absence of PRC2 activity . Taken together these results substantiate the relation between PRC1 and Rest in mES cells and suggest that Rest-PRC1 can be recruited to neuronal genes independently of the PRC2 complex . Finally , we wanted to study the interplay between REST and the PcG complexes at genes that are induced during differentiation . In order to do this we took advantage of the fact that NT2-D1 cells can be induced to differentiate along the neuronal path by retinoic acid ( RA ) - stimulation , and used these cells to study the dynamics of REST and RNF2 binding on a number of selected target genes that were co-regulated by REST and RNF2 ( Figure 2B ) . RA-stimulation for 3 days led to an almost complete loss of the pluripotency factor OCT4 ( Figure 7A ) , while the protein levels of EZH2 , RNF2 and REST were only marginally affected . Furthermore , OCT4 and another pluripotency factor NANOG were almost completely silenced transcriptionally ( Figure 7B ) , while the REST transcript was unaffected and HOXA1 was strongly induced ( Figure 7B ) . The genes we selected for ChIP analysis were all induced by RA-stimulation ( Figure 7D and 7F ) , but can be divided into two groups , based on the events taking place at the REST binding sites . The first group is represented by MEIS1 and TBX3 , which both have reduced REST binding to the gene after 3 days of RA-stimulation ( Figure 7C ) . These genes lost RNF2 binding and had reduced H3K27Me3 after RA treatment . For the second group of genes represented by MEIS2 and DLX5 , REST stayed on the genes after 3 days of RA-stimulation , while RNF2 was lost ( Figure 7E ) . These genes also lost H3K27Me3 ( MEIS2 ) or had significantly reduced H3K27Me3 after RA-stimulation ( DLX5 ) . The data suggested , that gene induction in response to RA-stimulation does not necessary require displacement of REST , but correlated with the displacement of PRC1 ( RNF2 ) and reduction in H3K27Me3 . The determining factor ( s ) for how REST-PcG complexes are regulated at specific target genes is still an open question . We imagine that post-translational modifications of REST and/or PcG proteins could be part of a mechanism regulating the interaction ( s ) between these factors and the affinity of REST for DNA . Furthermore , other REST associated factors , beside the PcG proteins , could affect the affinity of REST complexes for chromatin on individual genomic loci .
In Drosophila there are at least 5 different transcription factors shown to be important for PcG recruitment [32] , however , among these only the ortholog of Pho called YY1 is preserved in mammals . Even though YY1 and a second transcription factor , E2F6 , have been found to interact with PRC2 [36] , [46] and PRC1 in mammals [13] , [14] , [35] , [47] , there have been only two studies presenting evidence to support the existence of mammalian PREs . The first potential mammalian PRE identified was an element shown to regulate the expression of a hindbrain segmentation gene MafB ( Kreisler ) . This putative PRE was shown to recruit PRC1 and PRC2 , thereby mediating silencing of an ectopically introduced transgene in both flies and mice [48] . In a second study Woo et al . found a 1 . 8 Kb element between the HOXD11 and HOXD12 genes that was regulated by PcG proteins and H3K27Me3 [49] . Interestingly , this locus contained several YY1 binding sites and YY1 binding to the locus was coincident with PcG enrichment . Loss of YY1 binding sites had however only modest effect on repression . Based on previous studies showing H3K27Me3 enrichment around REST binding sites ( RE1 elements ) in human T-cells [50] and the fact that we identified REST in the purification of the CBX8 interacting protein HAN11 ( [44] and Figure S1 ) , we decided to investigate the functional relationship between REST and the PRC1- and PRC2-complexes . We found that REST was in complex with core members of both the PRC1- and PRC2-complexes in the human NT2-D1 , HEK293 cells and mouse embryonic stem cells . For PRC2 we found that REST interacted with the three essential core members: EZH2 , SUZ12 and EED , showing that REST was associated to a functional PRC2 complex . Due to the existence of both PRC1 and other RNF2 containing complexes , we performed a more detailed analysis of the REST-PRC1 complex . This analysis revealed that REST form complex with RNF2 and other selected core members of the PRC1 complex ( BMI1 , NSPC1 , CBX7 and CBX8 ) . Importantly , we did not detect E2F6 , HP1γ or BCOR in the REST IPs , proteins that have previously been described in RNF2 containing E2F6 . com-1 ( E2F6 and HP1γ ) [13] and Fbxl10-BcoR ( BcoR and HP1γ ) [14] , [37] complexes . These observations suggest that the REST-PRC1 complex that we have identified is distinct from the previously described Rnf2-containing complexes . Recently , an interesting work related to PRC2 and Rest was published by Tsai et al . [27] . They found that the previously described long non-coding RNA , HOTAIR [51] , [52] interacted not only with PRC2 through its 5′end , but even bound a LSD1/CoREST/REST-complex through the 3′end and in this fashion function as a modular bifunctional RNA . To understand if HOTAIR or other ncRNAs could be involved in bridging PRC1 and PRC2 to REST in the complexes that we identified , we treated our REST immunoprecipitates from NT2-D1 cells with RNases , but did not observe changes in the relative amounts of PRC1 and PRC2 subunits bound to REST . Thus , we conclude that the REST-PcG complexes that we have identified are stable in the absence of ncRNAs such as HOTAIR . Further studies are needed to determine if PRC1 and PRC2 are part of one and the same REST-PcG complex or are separate entities . In support of a functional interaction between REST and PRC1 , we found a significant number of genes that were co-regulated in the NT2-D1 cells . In line with REST and PRC1 being repressors of gene activity , the majority of co-regulated genes were up-regulated ( 258 ) and gene-ontology analysis revealed an enrichment of genes involved in development and neuronal function ( Table S9 ) . This is in agreement with previous publications showing that REST and PRC1 directly target genes involved in developmental processes and neuronal function . In contrast , the group of co-down-regulated genes did not enrich for any particular biological function suggesting that this group of genes were not primary targets of REST and PRC1 . In our genome-wide analysis of Rest and PcG occupancy in mES cells , we furthermore found that both PRC1 ( Rnf2 ) and PRC2 ( Suz12 , Jarid2 ) subunits specifically enriched at Rest binding sites . Similarly to the co-up-regulated genes identified in NT2-D1 cells , there was enrichment for genes involved in developmental and neuronal function among the genes targeted by both Rest and the Polycomb complexes ( Figure S5 and Table S7 ) suggesting a similar set of target genes in these different cell types of human ( NT2-D1 cells ) and mouse origin ( mES cells ) . However , the loss of REST in NT2-D1 and mES cells have different outcome with respect to transcription . Whereas , loss of REST in NT2-D1 cells leads to up-regulation of a large group of genes ( 1862 ) the effect in mES cells was less than expected and did not compare to the number of target genes [41] ( Tables S1 and S2 ) . This suggest that genes bound by REST in NT2-D1 and mES cells are differently regulated and that in mES cells , Rest target genes are regulated by additional repressive mechanisms , which ensure that developmental genes are kept in an “off-state” . The genome-wide analyses for PcG occupancy in mES cells , demonstrated a specific enrichment of the average Rnf2 , Suz12 and Jarid2 signals at the positions corresponding to Rest binding sites ( Figure 4A and Figure 6B ) . To improve the validity of these observations , we included a range of controls and extra analyses . Bona fide signals may co-exist at specific genes or gene features in a manner that seems significant due to congregations at groups of genes or gene features . Therefore , we used a set of randomly chosen control positions that matched the Rest-binding sites in terms of TSS-proximity . In comparison , Rest binding sites scored positive for PcG proteins ( Rnf2 , Suz12 and Jarid2 ) 4–5 times as frequent as the matched control regions ( Figure 4D ) . Co-occurrence analysis of Rest and Rnf2 furthermore allowed us to rule out that Rest and Rnf2 co-enriched by chance because both proteins were frequently found at TSS ( Figure 6F ) . With these tests , we are confident that the co-localization of Rest and PcG complexes in the genome was unlikely to be a mere coincidence . In support of this we furthermore found that Rest was required for PcG protein binding to a highly significant number of Rest binding sites in mES cells as well as for the binding to a selected number of neuronal target genes in NT2-D1 cells . Taken together our observations , strongly support a role for Rest as a DNA specific PcG recruitment factor . To understand if PRC2 was required for Rest-mediated PRC1 binding , we took advantage of the Eed−/− mES cells , which lack PRC2 activity and therefore the H3K27Me3 mark [53] . Accordingly , PRC1 , which is recruited to H3K27Me3 marked chromatin regions through the chromodomain-containing CBX proteins , was expected to be displaced from chromatin in cells lacking Eed . Nonetheless , when Eed−/− mES cells were compared to Wt mES cells , the Rnf2 signal strength was only marginally affected at Rest binding sites , although roughly 9 out of 10 Rnf2-binding sites were lost in Eed−/− mES cells ( Figure 6A and 6B ) . This interesting finding demonstrated that Rest-dependent recruitment of PRC1 to chromatin occured independently of PRC2 activity and the H3K27Me3 mark , and offers a mechanism that , in part , could explain why Wt and Eed−/− mES cells have similar levels of the PRC1 catalyzed H2AUbi mark [19] . Given the aforesaid recruitment of PcGs to Rest binding sites and our biochemical data showing interaction between Rest , PRC2 and PRC1 , we were highly puzzled by the observation that the absence of Rest , at a subpopulation of Rest binding sites , resulted in an increased binding of both PRC1 and PRC2 members in the Rest−/− mES cells . Interestingly , comparisons with matched control regions , revealed that both gain and loss of PcG binding in the Rest−/− mES cells were highly significant ( except Suz12 loss ) and specific to Rest binding sites . This suggested that the changes in PcG binding between Wt and Rest−/− mES cells indeed was a result of the absence of Rest and that Rest recruited PcG proteins to some genomic loci , while limiting the binding to others . In agreement with the view that PRC1 is recruited downstream of PRC2 , the increase in Rnf2 binding at some Rest binding sites were linked to a similar increase in Suz12 binding at these sites ( Figure S3 ) . Moreover , when the influence of the position relative to TSS or CpG-islands was compared to that of PRC2 levels , it was clear that the gain in Rnf2 binding observed at Rest binding sites , close to TSS and CpG-islands , were indirect consequences of increased PRC2 recruitment to these entities ( Figure 5E ) . Based on this , and the genome-wide analyses of Rnf2 in Eed−/− mES cells , we propose that PRC2 is dispensable for the majority of PRC1 recruitment far from CpG-islands , whereas PRC2 is the prime mechanism responsible for CpG-island proximal PRC1-recruitment , and only a minority of the CpG-island proximal Rnf2 binding sites is instead depending on alternative mechanisms , such as Rest . The observed gain of PRC2-signal at many Rest binding sites in the Rest−/− mES cells , suggest that Rest not only serves as a factor recruiting PcG complexes , but can directly or indirectly limit the amounts of PcG proteins on chromatin near some of its binding sites . Indeed , Rest is known to interact with other co-repressors ( HDACs , CoRest/Lsd1 and G9a ) , which may affect chromatin structure in a way that limit PRC2 binding . It is tempting to speculate that this is part of a mechanism , which has evolved to increase robustness in gene regulation by compensating for the loss of one repressive complex , and thereby ensure redundancy and fidelity in the silencing of lineage specific genes . This might also be the reason for the relatively modest effects of Rest knock-out or knock-down on gene expression in mES cells [41] compared to the more pronounced changes observed in the NT2-D1 cells ( Figure 2B ) . In line with this , using the expression analyses data from Færk et al . [41] , looking at genes with either gain or loss of Rnf2 or Suz12 from our ChIP-seq analysis , we found no significant effects on gene expression correlating with the change in PcG binding in mES cells ( Figure S6 ) , although these genes are regulated by PcG according to Leeb et al . ( Figure S7 ) . During the final preparation of our work a study from the Kerppola laboratory was published [54] , which supports our findings of an interaction between PRC1 and Rest . Based on a selected number of genes , divided into either promoter proximal- or promoter distal-Rest binding sites , they concluded that Rest inhibits PRC1 recruitment at proximal binding sites , while recruiting PRC1 at distal binding sites . Since promoter proximal elements are rich in CpG islands this conclusion is in agreement with our genome-wide analysis of PcG protein binding and annotated CpG islands , which showed that Rest is required for PRC1 binding in the absence of CpG islands . However , it should be noted that whereas the Ren et al . study , on the basis of relatively few selected target genes , gives the impression that all Rest binding sites co-localize with PRC1 , our genome wide analysis show direct overlap of Rest and PRC1 on 165 Rest binding sites out of a total of 3 , 378 Rest binding sites identified in our ChIP-sequencing analysis in mES Wt cells . Of these 165 co-localizing sites , 80 were found outside promoter proximal CpG islands and 85 on CpG islands . Furthermore , our study revealed that the increase of PRC1 ( Rnf2 ) at promoter proximal Rest binding sites correlated with an increased recruitment of PRC2 to CpG islands at these locations . The increase in PRC2 was not directly translated into increased H3K27Me3 as one might expect , which suggest that H3K27Me3 in those regions were already quite high . Therefore , it is possible that the recruitment of PRC2 to these CpG rich regions is controlled by other histone modifications mediated by Rest complexes , which prevent efficient binding of PRC2 to the H3K27Me3 marked regions [55] . Furthermore , related to the fact that only about 5% of the Rest binding sites in mES cells showed co-localization with Rnf2 , although PRC1 complexes are very abundant , suggest that other factors have influence on the recruitment of the Rest-PRC1 complex . Even though , we found that the Rest-PcG protein complexes were biochemically stable in the absence of ncRNAs ( RNase treatment ) it is still possible that ncRNAs such as HOTAIR could play a significant role in stabilizing Rest-PcG complexes on chromatin and thereby influence target gene specificity . Considering that Rest binding sites might be part of mammalian PRE elements , it is furthermore likely that other transcription factors , binding in the vicinity of Rest and with affinity for PcG complexes , influence whether or not PRC1 is stably interacting with REST on a particular binding site . Since our HAN11 double-tag affinity purification revealed a number of other transcription factors beside REST , future studies will aim at investigating if any of those can co-operate with REST in targeting PcG complexes to specific genomic loci . In conclusion , we have shown that the transcription factor REST interacts with PRC1- and PRC2-complexes , interactions that we found to be independent of ncRNAs . Importantly , our data furthermore showed that the PRC1 complex can be recruited to a number of Rest binding sites independently of PRC2 activity and CpG islands . Surprisingly , we also found that there exist a CpG-island-associated increased recruitment of PRC2 in the absence of Rest , at a number of genes . We propose that this up-regulation of PRC2 binding functions to prevent unscheduled activation of differentiation specific genes and contribute to the robustness of mES cells . To understand the details regarding this compensatory mechanism and whether genomic regions , recruiting REST-PRC1 independently of PRC2 activity , are part of mammalian Polycomb Responsive Elements ( PREs ) will be the focus of future experiments .
Mouse embryonic stem ( mES ) cells were cultured on 0 . 1% ( w/v ) gelatin-coated plates in Glasgow medium ( Sigma ) supplemented with glutamax-1 ( Gibco ) , non-essential amino acids ( Gibco ) , 50 µM 2-mercaptoethanol–PBS , 15% ES-cell-qualified FBS ( Gibco ) in the presence of 1 , 000 U/ml of LIF ( Millipore ) , and 1% ( v/v ) pen/strep . In addition , the Rest−/− and Wt control mES cells were cultured in the presence of feeder cells ( Mitomycin C treated primary mouse embryonic fibroblasts ) . The mES Eed−/− and Rnf2−/− mES cells were provided by Dr . Anton Wutz . The Rest−/− and Wt control mES cells were provided by Dr . Zhou-Feng Chen and Dr . Helle Færk Jørgensen . HEK293FT and NT2-D1 cells were grown in DMEM ( 4 . 5 g/l D-glucose , Gibco ) supplemented with 10% ( v/v ) FCS and 1% ( v/v ) pen/strep . To induce neuronal differentiation of NT2-D1 cells , exponentially growing cells were seeded at 30% confluency and 24 hours later retinoic acid ( RA ) was added to a final concentration of 10 µM . Medium was changed every second day . Cells or nuclear preparations were lysed in high-salt ( HS ) lysis buffer ( 50 mM Tris , pH 7 . 2 , 300 mM NaCl , 0 . 5% ( w/v ) Igepal CA-630 , Leupeptin ( 1 µg/ml ) , Aprotinin ( 1 µg/ml ) , 1 mM PMSF , 1 mM EDTA ( pH 7 . 4 ) ) . After a short 2 second ultrasonication pulse ( Branson Sonifier; 20% max amplitude ) the lysates were left on ice for 30 min . After centrifugation at 20 , 000 g ( 4°C ) for 15 min and 100 , 000 g for 30 min the lysates were passed through a 0 . 45 µm low protein binding filter ( Ultrafree MC spinfilter , Millipore ) followed by a 0 . 22 µm filter before the protein concentration was determined by Bradford assay ( BioRad ) . Between 5 and 10 mg of protein was loaded on a Superose 6 HR 10/300 ( 24 ml ) equilibrated in GF buffer ( 25 mM Tris , pH 7 . 2 , 150 mM NaCl , 0 . 2% ( w/v ) Igepal CA-630 , 1 mM EDTA ( pH 8 . 0 ) , 1% ( w/v ) glycerol ) in 1 ml ( flowrate 0 . 3 ml/min ) . One ml fractions were collected ( 4°C ) and characterized by Western blotting and afterwards processed for immunoprecipitation by pooling peak fractions of interest . Typically 100–200 µl ( pooled fractions ) was used for each immunoprecipitation using anti-REST or control IgG . For direct Western blotting , to visualize the elution profile of individual proteins , an equal volume of each fraction was mixed with 2X LSB and heated at 95°C for 5 min . Ten µl was loaded per lane for SDS-PAGE and Western blotting . All SDS-PAGE gels used were precast 4–12% gradient gels or 10% homogenous gels ( Invitrogen ) using the MES buffer system . Anti-REST and general IgG ( rabbit; DAKO ) were pre-coupled to proteinA-Sepharose beads ( 0 . 5 µg antibody per 40 µl 1∶1 slurry ) and cross-linked using DMP ( dimethyl pimelimidate dihydrochloride , Fluka ) at pH 9 . 0 in 200 mM borate buffer according to standard procedures . For samples treated with RNase V1 and RNase A the immunoprecipitates were washed twice in HS buffer followed by two washes in HS buffer adjusted to 500 mM NaCl and ones in HS buffer before adding 0 . 1 Unit of RNase V1 ( Ambion ) and 0 . 1 Unit of RNase A ( Roche ) in HS buffer . Samples were incubated at 10°C for 30 min followed by one wash in HS buffer , one wash in 500 mM HS buffer and one final wash in HS buffer . Samples were eluted in 2X LSB and heated to 95°C for 5 min before SDS-PAGE and immunoblotting . RNA was purified using RNeasy Plus Mini kit ( Qiagen ) and cDNA was generated by RT–PCR . Quantifications were done using the Fast SYBR Green Master Mix ( Applied Biosystems ) and an ABI Step One Plus . Beta-Actin was used for normalization . The sequences of the primers used can be found in Table S8 . Cells were fixed for 10 min in culture media containing 1% formaldehyde and were processed for ChIP as previously described [44] . Antibodies used: Rabbit IgG ( DAKO ) , rabbit anti-H3K27Me3 ( Cell Signalling , C36B11 ) , rabbit anti-H3 ( GERA; antigen sequence: CGIQLARRIRGERA ) , rabbit anti-REST ( Millipore , 07-579 ) , rabbit anti-RNF2 ( NAST; antigen sequence: NASTHSNQEAGPSNKRTKT ) , rabbit anti-SUZ12 ( Cell signaling ) , rabbit anti-Jarid2 ( Novus Biologicals ) , anti-CBX7 ( “RELF” , [44] ) , anti-CBX8 ( “LAST” , [44] ) , anti-NSPC1 ( XW5 ) . Western blotting was performed according to standard procedures using the following antibodies: Anti-RNF2 ( “NAST” ) , anti-REST ( Millipore , 07-579 ) , anti-EZH2 ( BD43-43 ) , anti-OCT4 ( ab19857 , Abcam ) , anti-GAPDH ( 6G5 , Biogenesis ) , anti-TUBULIN ( Sigma , T6074 ) , goat anti-SUZ12 ( Santa Cruz , sc-46264 ) , anti-BCOR ( Novus Biologicals , NB100-87005 ) , anti-CBX7 ( “RELF” , [44] ) , anti-CBX8 ( “LAST” , [44] ) , anti-NSPC1 ( XW5 ) , anti-E2F6 ( TFE61 ) , anti-BMI1 ( DC9 ) , anti-EED ( AA19 ) . Blots were developed using HRP-conjugated anti-rabbit , mouse or goat antibodies , depending on the species of the primary antibody , and enhanced chemiluminescence ( ECL; Pierce ) . All exposures were done using Hyperfilm ( Amersham ) . VSV-G virus particles were generated by calcium phosphate-mediated co-transfection of PAX8 and VSV-G plasmids together with the pLKO . 1 targeting construct in 293FT cells . pLKO . 1 targeting constructs: pLKO . 1-RNF2 ( TRCN0000033697 , Sigma ) , pLKO . 1-REST ( TRCN0000014783 , Sigma ) and pLKO . 1-RCOR1 ( TRCN0000147184 ) . An empty pLKO . 1 vector was used as control . NT2-D1 cells were incubated with virus supernatant for 8 hours and selection using 2 µg/ml puromycin started after 24 hrs PI . The NT2-D1 cells were harvested for ChIP , WB , and RNA after 4 days of selection . Total RNA was extracted from NT2-D1 cells treated with pLKO . 1-Ct or shRNA to knockdown REST and RNF2 mRNA . RNA was prepared from 3 independent experiments . 300 ng of total RNA from each experiments was processed for microarray expression analysis according to Affymetrix standard procedures . Up-regulated genes were identified using Microsoft Excel 2003 by individual probe-sets showing more than a two-fold change compared to control and with p-values below 0 . 05 in Student's t-test . Cluster analysis was performed using BRB-ArrayTools v3 . 81 ( http://linus . nci . nih . gov/BRB-ArrayTools . html ) using standard settings . Before cluster analysis probe sets were first filtered for 1 ) at least one array with a read above 10 and 2 ) more than one array deviating at least 1 . 5 fold from average . DNA from three parallel ChIPs were pooled and 10 ng was used for making ChIP-seq libraries . Libraries were generated according to Illumina recommendations and sequencing was done on a Genome Analyzer II ( Illumina ) . High quality reads ( Chastity score > = 0 . 6 ) were aligned to the mouse genome ( mm9 ) using Eland ( Illumina ) allowing up to two mismatches within the first 32 bases . Reads not aligning uniquely to the mouse genome were removed and profiles were presented using the UCSC Genome Browser ( http://genome . ucsc . edu/ ) [56] . For detailed information about data handling and analyses see Procedure S1 . | Multicellular organisms are composed of a large number of specialized cell types that all originate from the Embryonic Stem cell ( ES cell ) . It is crucial for the maintenance of naïve ES cells that developmental genes are kept in an off-state until appropriate differentiation stimuli are received . Polycomb Repressive Complexes , PRC1 and PRC2 , are bound at and repress the activity of a large number of key developmental genes in ES cells and at different stages of differentiation . While in Drosophila the PRC complexes are recruited to DNA elements called Polycomb Response Elements ( PREs ) , through the interaction with transcription factors; examples of such factors remain poorly characterized in mammals . We here demonstrate that the transcription factor Rest interacts with and is required for recruitment of PRC1 and PRC2 to a subset of Rest target genes in mouse embryonic stem ( mES ) cells . In line with REST being a repressor of neuronal genes , we found that PRC1 and PRC2 co-localized with REST at genes involved in neuronal development and got displaced during neuronal differentiation . Based on our data we propose that the PRC1 and PRC2 complexes function as co-repressors for Rest to control the timed expression of developmental genes in the process of cellular differentiation . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"developmental",
"biology",
"biology",
"genomics",
"molecular",
"cell",
"biology",
"genetics",
"and",
"genomics"
] | 2012 | REST–Mediated Recruitment of Polycomb Repressor Complexes in Mammalian Cells |
The hair cycle is a dynamic process where follicles repeatedly move through phases of growth , retraction , and relative quiescence . This process is an example of temporal and spatial biological complexity . Understanding of the hair cycle and its regulation would shed light on many other complex systems relevant to biological and medical research . Currently , a systematic characterization of gene expression and summarization within the context of a mathematical model is not yet available . Given the cyclic nature of the hair cycle , we felt it was important to consider a subset of genes with periodic expression . To this end , we combined several mathematical approaches with high-throughput , whole mouse skin , mRNA expression data to characterize aspects of the dynamics and the possible cell populations corresponding to potentially periodic patterns . In particular two gene clusters , demonstrating properties of out-of-phase synchronized expression , were identified . A mean field , phase coupled oscillator model was shown to quantitatively recapitulate the synchronization observed in the data . Furthermore , we found only one configuration of positive-negative coupling to be dynamically stable , which provided insight on general features of the regulation . Subsequent bifurcation analysis was able to identify and describe alternate states based on perturbation of system parameters . A 2-population mixture model and cell type enrichment was used to associate the two gene clusters to features of background mesenchymal populations and rapidly expanding follicular epithelial cells . Distinct timing and localization of expression was also shown by RNA and protein imaging for representative genes . Taken together , the evidence suggests that synchronization between expanding epithelial and background mesenchymal cells may be maintained , in part , by inhibitory regulation , and potential mediators of this regulation were identified . Furthermore , the model suggests that impairing this negative regulation will drive a bifurcation which may represent transition into a pathological state such as hair miniaturization .
The miniorgan of the hair follicle represents a complex biological system that undergoes repeated phases of death and regeneration over its lifetime [1]–[3] . Understanding of the hair cycle and its regulation would shed light on many other complex systems relevant to biological and medical research including morphogenesis , stem cell biology , response to environmental perturbations and general spatiotemporal patterning [4] . The stages of the hair cycle have been well documented , at least from a morphological standpoint , in mouse models [5] . The period of hair growth , known as anagen , involves rapid proliferation of follicular epithelial cells , such as MatriX ( MX ) cells in the hair bulb , which surround a key group of mesenchymal cells that form the dermal papilla ( DP ) . Matrix cells differentiate to eventually compose various epithelial populations of the hair shaft . Anagen is followed by catagen , which is characterized by high levels of apoptosis . Finally , telogen is typically described as a quiescent period between growth phases . The molecular mechanisms underlying this cyclical pattern of death and renewal in hair follicles are not well understood; however , some general concepts , as well as specific molecular regulators , have been identified . One key aspect is the communication between epithelial and mesenchymal cells . Numerous studies have identified physical interactions between these cell populations , as well as several possible signaling molecules [6] . One well studied signaling molecule of the hair cycle is Tgfβ2 , which is synthesized and secreted by DP cells . The evidence suggests that , in general , Tgfβ2 suppresses proliferation and induces catagen-like changes in the follicle , including apoptosis of MX cells [7] . However , recent studies have identified a Tgfβ2 mediated pathway which activates epithelial stem cells to promote hair follicle regeneration [8] . This underscores the complexity of the signaling pathways involved . Mathematical models of general features of hair cycling have also been studied . In a recent study by Murray et al . , the authors model follicle growth and coupling as an excitable medium [9] . Their model incorporates general aspects of hair cycle regulation , and shows qualitative agreement to experimental observations . Also recently , Al-Nuaimi et al . developed a general model for hair cycling based on observations in the literature [10] . These authors derived a mathematical , kinetic model which proposed that negative feedback between dynamic MX keratinocytes and static DP cells could reproduce the cyclical growth patterns of the hair follicle . Although these models are significant , they do not attempt to incorporate any specific molecular details in a data-driven approach by formally analyzing large scale experimental data sets . In the study by Lin et al . , mRNA microarrays were compiled over the first three rounds of hair growth: morphogenesis , the second naturally synchronized cycle and a depletion-induced cycle [11] . The results demonstrated recurrent gene expression corresponding to hair growth , and the authors specifically focused on genes related to circadian rhythms . However , the study does not address the many other genes observed to have similar patterns . Currently , an unmet need is the development of data-driven approaches that can couple the existing transcriptome-wide data to systems-level properties of hair cycling using formal dynamic models . Nonlinear dynamical models have provided valuable insights into many oscillating biological systems [12]–[14] , and have even been used to suggest general design principles of oscillating metabolic and signaling networks [15] . In general , simplified oscillator models have been developed to describe properties resulting from oscillator interactions or coupling . One such model is the well-studied Kuramoto model [16] . Here all oscillators are interconnected with the same coupling strength , and studied as a single mean field . Although a major simplification , mean field models have successfully described high level properties of many large , complex systems including statistical mechanics ( for a review see [17] ) , economics [18] , [19] and even social networks [20] . Importantly , the Kuramoto model is capable of capturing a critical phase transition from an incoherent state to one in which all oscillators converge to a single , coherent cluster . This behavior is referred to as synchronization and is quantified by complex order parameters [21] . Modified and extended versions of the Kuramoto model have been used in many complex systems [22] including synthetic genetic networks [23]; cyclical gene expression and cellular networks [24]; neural networks for memory and brain activity [25] , [26]; chemical oscillators [16]; and laser arrays [27] , [28] . We would not expect such models to be capable of incorporating or identifying mechanistic molecular interactions or specific details; however , given the above literature evidence , they can be quite successful at describing systems level properties , such as synchronization , and the sufficient , underlying conditions that can produce them . Our aim here was to investigate a subset of genes whose expression changes as a function of time in a potentially periodic manner , similar to the cyclical nature of hair growth . Previous modeling studies , which have focused on general aspects of hair growth , represent important initial steps in applying mathematical strategies to understanding the hair cycle [9] , [10]; however , these models are not driven by molecular-level data . In contrast , other studies use high-throughput molecular-level data to identify important targets , and they apply additional experiments to delineate specific molecular mechanisms [11]; however , these studies are limited to investigation of a small number of genes , and they do not attempt to place the observations into a quantitative modeling context . In this study , we focused on two complementary mathematical modeling strategies that look at high-level features , such as average dynamic behavior , that is based on the individual patterns of thousands of genes . Thus , we are attempting to bridge the gap between the two strategies described above . Using whole skin , transcriptome-wide expression data , we demonstrate the existence of two subsets of genes that have synchronized , out-of-phase expression profiles . Motivated by this observation , we applied a coupled oscillator modeling framework to identify a specific coupling configuration that spontaneously , and stably reproduced the observed synchronization . We then applied a 2-population mixture model to associate the corresponding gene clusters to two computationally determined populations , a rapidly expanding population and a relatively static background population . The estimated population dynamics indicated an association between computationally derived background/expanding populations and the mesenchymal/follicular epithelial cells , respectively . Cell type specific enrichment analysis and experimental imaging with in situ hybridization and immunofluorescence all demonstrated similar associations . The results describe a coupling scheme , between these two cell populations , which would be sufficient to maintain the observed synchronization . Specific signaling molecules were also identified as being priority follow-up targets for drivers of synchronization . To our knowledge this is the first attempt at integrating high-throughput molecular data with a mathematical model to predict systems level properties , such as synchronization and population dynamics .
Given the proposed cyclical nature of hair growth , we investigated the possibility of periodically expressed mRNA in the microarray data collected by Lin et al . [11] . We assumed that such expression patterns may relate to hair cycle regulation . We applied a periodic identification scheme for non-uniformly sampled data [29] . This method estimates a discrete Fourier Series Decomposition ( FSD ) for each expression signal by robust regression . We identified 4627 probesets ( mapping to 3567 unique genes ) as significantly periodic signals with a false discovery rate of 10% . Using the semantic measure Normalized Google Distance ( NGD ) , we found that 315 of the corresponding periodic genes had a notable proximity to the hair cycle described in a survey of PubMed abstracts . This translates to an enrichment p-value of 2 . 5E-5 . For example , periodic genes with the lowest NGD are discussed in the literature as being related to hair pigmentation ( Mc1r , Tyrp1 , Stx17 ) , growth and cycle regulation ( Liph , Foxn1 ) , disorders and malformations ( Lpar6 , Zdhhc13 , Krt85 ) and general associations to hair ( Krt28 ) . For a full list of genes and related NGDs to hair see Supplementary File S1 . To further investigate the periodic expression , we assigned a specific frequency and a phase shift to each signal . This was done using the Principal Periodic Component ( PPC ) as an approximation to the FSD . Both the PPC and FSD reasonably recapitulated the time course trajectories , primarily the low frequency expression signals ( Supplementary Figure S1 top ) . Furthermore , the majority of the periodic signals , 3988 probesets , were associated to this low frequency , which corresponds to a period of 31 days ( Supplementary Figure S1 bottom ) . The 31 day period of expression was on the same time scale as the hair cycle , and further suggested a relationship between the corresponding genes and hair growth and regulation . Although repeating cycles were not directly observed to demonstrate cyclical behavior , we note that the data was a composite of both the second natural and depilation-induced hair growth cycles and , therefore , the expression patterns were common to , at least , these two cycles . For visual examination , we sorted the periodically expressed probesets by frequency and phase shift ( Figure 1A ) . We compared this data to the PPC trajectories ( Figure 1B ) to underscore the similarities . We noted a distinct clustering of the expression signals , including two clusters within the low frequency probesets . The lower and upper clusters demonstrate maximal and minimal expression near the end of anagen , respectively . This reciprocal periodic behavior is referred to as out-of-phase periodicity . We calculated the phase shift based on the time to the next maximum value . The phase shift ( Figure 1C ) clearly identifies these two tightly clustered groups . We associated 1452 probesets to cluster one and 2536 to cluster two . Cluster one and two correspond to probesets that are predicted to reach maximum expression at approximately 33 ( first anagen phase after morphogenesis ) and 48 ( following telogen phase ) days postnatal , respectively . The separation of the phase shift further indicates that the two groups are almost exactly out-of-phase . The mean of the two groups is separated by 15 . 4 days which in polar coordinates corresponds to approximately 180° . Although fast cycling genes were identified , we chose to overlook this group due to a relatively poor fit to the PPC , median coefficient of determination was less than 0 . 5 ( Supplementary Figure S1 Top ) . However , given that the data was derived from full-thickness mouse skin , it is possible that cyclic gene expression in Keratinocytes could be contributing to the short period signal . This possibility was strengthened with the recent report that human epidermal stem cell functions are regulated by circadian oscillations [30] with a period of 24 hours in vitro . Additional experiments , with higher time-resolution sampling , may better describe fast cycling genes and may provide a link between Keratinocytes , hair-cycling and circadian rhythms . The clustering of periodic signals was quantified using complex order parameters [21] , [31] , [32] . Considering only the periodic component of the low frequency expression signals corresponding to a 31 day period , we can visualize the expression as points moving around the unit circle in the complex plane as they travel through the cycles . How tightly grouped these points are can be quantified by a set of order parameters , : ( 1 ) where is the number of oscillators and is the instantaneous phase or the position of oscillator on the unit circle at time . See Methods for more details on the formulation of EQ 1 . When studying systems of coupled oscillators , the magnitude of , denoted , is used to quantify the coherence or synchronization of the system ( Supplementary Figure S2 shows typical values for specific configurations of points ) . In such systems , a high level of synchronization is typically the result of coupling between oscillators . The low frequency expression signals in the hair cycle demonstrated high out-of-phase synchronization as measured by , as well as a notable asymmetry , due to uneven sized clusters , measured by ( Figure 2A ) . As a negative control , we randomized the oscillator phases to demonstrate and near zero for a similar , but un-clustered system ( Figure 2A , black lines ) . Synchronization was further exemplified by considering order parameters calculated for the specific clusters , denoted and ( Figure 2B , green and blue lines corresponding to clusters shown in Figure 1C ) . This level of synchronization appears to be dynamically stable throughout the time course . If we can expect similar molecular behavior underlying subsequent cycles of hair growth , we would anticipate these periodic expression signals to repeat . In an ideal case the low frequency gene expression could then be viewed much like a system of oscillators , and the observed dynamic stability could be investigated in that context . We believed that strong synchronization over multiple expression signals was indicative of regulation between the corresponding genes . As mentioned above , systems of multiple agents with periodic behavior are often described using coupled oscillator models . Furthermore , and most significantly , we found a striking similarity between the observed expression in the mouse hair cycle and a simple system of coupled oscillators formulated by Hong and Strogatz [33] . In this model one possible attractor ( or long time behavior ) was the synchronization of two , asymmetric , out-of-phase , oscillator clusters . Spontaneous , stable synchronization was observed when one cluster was positively coupled to the system's macroscopic rhythm , embodied by complex order parameter ( see EQ 1 ) , and the other was negatively coupled . In this case positive or negative coupling indicates oscillators that move towards or away from , respectively . Although such models are a significant simplification from the true biology , the qualitative agreement was encouraging , and we wished to investigate if reasonable insights could be drawn from such an abstraction . In the following we consider a system of Low Frequency Oscillators ( LFO ) corresponding to the 31 day period expression signals identified above . We first considered if our system could be quantitatively described by the above model . At the level of individual oscillators this model can be written as ( 2 ) where is the phase of the th oscillator , is the natural or intrinsic frequency for , is the coupling constant , is the number of positively coupled oscillators , is the total number of oscillators , and from EQ 1 where the subscript is dropped for simplicity . The dot denotes change with respect to time . Here , we have assumed two groups representing positive and negative coupling denoted by superscripts + and − , respectively . The coupling constants are related by where , , . This model can be simplified to two dimensions when describing only the dynamics of the first order parameter for the two clusters ( recall Figure 1C , green and blue clusters ) , which was what we focused on here as a high level characterization of the system ( 3 ) where and are the first order parameters ( similar to EQ 1 with ) for the two oscillator groups related to positive and negative coupling , respectively; is the proportion of positively coupled oscillators and . The bar denotes complex conjugate . Given the first order parameters of our observed clusters ( and , recall Figure 2B green and blue lines ) and the number of oscillators for the two clusters ( which provides ) , we solved exactly for the relative coupling strength , , and the intrinsic frequency distribution , . We refer the reader to Methods for a detailed description of this process including additional assumptions . Interestingly , the only stable configuration was negative coupling of cluster 1 , the smaller cluster ( ) . Figure 3A shows that this configuration resulted in spontaneous , stable synchronization ( config 1 , red solid line , a corresponding movie of the individual oscillators is available in Supplementary File S2 ) , a steady-state magnitude of the first order parameter equivalent to the observed magnitude ( recall Figure 2A ) , and clusters that remain 180° out-of-phase ( upper panel of Figure 3A ) . Furthermore , if cluster 1 is positively coupled ( ) , the model system is unstable and no synchronization is observed ( Figure 3A purple dashed line , config 2 ) . The results suggest that a positive coupling of cluster 1 to the system is physically unlikely . This result provides us with our first biological insight , specifically , the genes corresponding to cluster 1 were negatively coupled to the system's macroscopic rhythm , and repelled by this average behavior of the system . If these genes can be associated with specific cell populations , then inhibition or repression of this population by the system would explain the simultaneous negative coupling of a large number of genes . Inhibition at the cellular level could be achieved by regulated apoptosis , which is consistent with a previous model of the hair cycle from the literature [10] . The authors described that negative feedback in the form of regulated apoptosis or inhibition of regulated proliferation could produce observed cyclical hair growth patterns . Using this coupled oscillator model , we investigated what other states could be possible if specific variables were changed . We constructed a bifurcation diagram that shows the steady-state behavior of the system given different values of ( see EQ 3 ) and assuming that other properties remain constant . Figure 3B shows three qualitatively different states for varying values of . Low values of relate to an incoherent state which has no synchronization and , therefore , no clustering . For large values of , we find the -state in which two clusters are stable , out-of-phase and oscillate with the same long period throughout a range of values , with increasing asymmetry to one cluster . Finally , we observed a traveling wave state for intermediate values of , here the period of the system is greatly reduced and even variable with respect to . These states were described by Hong and Strogatz [33] . Interestingly , the hair system lies at the edge of the -state , near a bifurcation into the traveling wave state . A reduction in corresponds to a decrease in the relative size of cluster 2 , and would reduce the effective negative coupling on oscillators in cluster 1 , which is what drives the system into a different state . Therefore the model anticipates that on average , removal of oscillators in cluster 2 would result in a loss of regulation which is specifically associated with varying , high frequency oscillations . Furthermore , we noted that the reduced period of the traveling wave state is similar to the fast cycling of short hairs in an existing model of hair cycle [10] , which was shown to result from changing parameters associated with negative feedback . This alternative state may be biologically related to the pathological state of hair miniaturization and androgenic alopecia . Although our model recapitulated the observed synchronization , we emphasize here the various abstractions introduced . First , we did not attempt to model the physical , molecular interactions involved , as we felt such an approach would be too error prone given limited data and a priori knowledge [34] . Instead , we chose to implement the simplest phenomenological model we could conceive to describe observed behaviors . Here oscillators represented probesets with periodic expression patterns ( see Methods EQ 7 ) . Modeled oscillators changed due to two terms: the intrinsic term , , and the coupling terms , ( both from EQ 2 ) . In a physical model the intrinsic term would represent some constant , external force that independently drives oscillations , in this case the intrinsic term represents all the unknown interactions that were the basis for periodic expression . As a result , insight and prediction related to these interactions was beyond this model's reach . In contrast the coupling terms were of most interest , and represented the average force felt by oscillators of a certain group , due to other oscillators . This coupling term is what drove the observed out-of-phase synchronization . The model was further abstracted as we specifically considered the average dynamics of the two clusters ( see EQ 3 ) , which were coupled via the average coupling of the underlying oscillators . In a sense the model treated interactions as being ‘smeared-out over all oscillators . Although exact physical interpretation is difficult , this average effect on the many corresponding genes may be the result of an extensive transcriptional program or a cellular event such as regulated apoptosis or proliferation , and suggests that these clusters may be related to specific cell populations . Given previous results , we hypothesized that the two observed clusters of LFO gene expression may relate to specific cell types or general cell populations . We noted that observed relative expression changes can be attributed to changes in relative cell population size as opposed to intracellular changes . Furthermore , opposite expression patterns can result when one population is constant in size and the other is changing ( See Supplementary Figure S3 , and Methods for details ) . To investigate further , we applied in silico microdissection [35] to estimate properties of two distinct cell populations from the heterogeneous hair cycle data . Here , the 2-population mixture model assumes static intracellular gene expression . Furthermore , we also assume one population is expanding , while the second is a static background population . We emphasize here that we make no assumption as to the specific cell types contributing to these populations nor the relative expression levels of genes within these populations . We then fit this model using all of 45k+ probesets from each micro array chip . From the heterogeneous hair cycle samples , we were able to approximate the dynamics of dominant expanding cell populations ( Figure 4 ) . The trajectory was consistent with cells associated with rapidly proliferating follicle epithelial cells . In particular the model estimated ( without any a priori knowledge ) a sharp and complete depletion of the expanding population within the catagen time frame . The model also identified differences between samples from the natural and induced cycle . Specifically , the model estimated a slower anagen onset in depletion induced mice , this was also observed by comparing morphologies of tissue sections in Lin et al . [11] . Furthermore , we demonstrated that these features were not estimated in a negative control ( see Supplementary Figure S4 ) , using a permutation strategy to simulate data with no hair cycle relationship . This suggested that our results were not due to an artifact or bias introduced in the analysis . These observations suggested that the in silico microdissection procedure was able to identify expanding cell populations compared to a static background populations . The model was also able to estimate static intracellular expression levels for each population . Combining this with the estimated population size , we were able to compare estimated expression levels to those observed in the data . Overall , we found that the majority of expression signals were not well described by such a simple model ( Figure 5A , blue histogram ) ; however , the model was able to describe the majority of the variation in the expression of the LFOs ( Figure 5A , pink histogram , 50% of probesets demonstrated a COD>0 . 5 ) . As above we did not identify such improved statistics of the LFOs in the permuted negative control . This suggested the significance of such a model in describing expression of the LFO subgroup . Using the estimated expression and standard error estimates , we identified probesets with statistically significant differential expression between the two estimated populations ( expanding and background ) in the form of a t-statistic . The t-statistic shows the expression difference relative to the estimated error . Again more statistically significant differential expression was found in the LFO subgroup ( Figure 5B ) . Probesets from the LFO subgroup were assigned to the population in which they demonstrated a statistically significant increase in expression , 99 . 7% of the LFO probesets met the statistical requirements for assignment . Figure 5C shows the remarkable similarity between clusters 1 and 2 ( recall Figure 1C ) and the expanding and background population respectively . In fact , dividing the LFO probesets in this manner yields the same division as achieved when considering oscillator phase ( Supplementary Figure S6 ) . Ultimately , this procedure allowed us to associate probesets and corresponding genes to two dynamically distinct populations . We next investigated the possibility that the estimated populations were associated to specific biological cell types involved in the hair cycle . We made use of two existing studies , Rendl et al . [36] and Greco et al . [37] , which dissected hair follicles into specific , predefined cell types . For each cell type , signature genes , genes expressed predominantly in that cell type , were identified . We then calculated the enrichment of these signature genes in the two estimated populations ( Supplementary Table S1 ) . In particular , we found a significant enrichment for epithelial MX and , to a much lower degree , ORS cells in the expanding population , and respectively , but not for the background population . Alternatively , we found a significant enrichment for mesenchymal DP cells in the background population , , but not for the expanding population . Other cell types , Melanocytes and Bulge Cells , were found to be significantly enriched in both populations , but to a larger degree in the background population ( see Supplementary Figure S7A for relative signature list size and overlap with estimated populations ) . We also observed overlap between the experimentally determined gene signatures , for example the overlap between Bulge [37] and DP [36] signature genes ( see Supplementary Figure S7B ) corresponds to an enrichment p-value . The results show that the computationally estimated populations do not uniquely represent a specific , predefined cell type; however , they do have distinct associations . Furthermore the association of epithelial MX and ORS cells to the rapidly expanding population and DP cells to the static background population was consistent with the known relative population dynamics of those cell types . We emphasize that DP population size does change throughout the hair cycle [38] . However , the DP population is not observed to undergo the enormous expansion and apoptosis characterizing hair epithelial cells [39]–[41] . Thus , conceptually the DP may be considered considerably less dynamic than the MX derived population [2] , which is consistent with our findings here . For a full list of significant probesets with the t-statistic indicating population association , and other metrics , see Supplementary File S3 . These results provide us with a second biological insight: the genes in LFO cluster 1 were associated with expanding cell populations of the follicle that were enriched for follicle epithelial cells as defined Rendl et al . [36] , and cluster 2 was associated with background populations that were enriched for mesenchymal DP cells . This bolsters our previous hypothesis that an inhibitory , possibly apoptotic mechanism , is acting on the expanding , epithelial cell population , and that this mechanism is involved in synchronized gene expression of the hair cycle . To verify specific gene expression patterns and to investigate the localization of gene products within the hair follicle , we applied qRT-PCR , In Situ Hybridization ( ISH ) , and protein antigen staining . To generate tissue samples , we aligned the hair cycle in 10-week old mice with the shave/depilatory induction protocol . Two animals were sacrificed for each of the five time points considered; however , qRT-PCR was done using four technical replicates for both samples and imaging shows results typical of multiple follicles observed over the two biological replicates . Phenotypic changes were quantified by melanogenesis scoring , which is known to correspond to active hair growth . Both biological replicates were observed to have entered anagen within 16 days of induction and to have completely finished the first round of post induction hair growth by 29 days , as determined by scores increasing from and then returning to zero ( Supplementary Figure S8A ) . Dorsal skin was sampled and prepared for RNA analysis or antigen staining at multiple time points throughout the cycle . An exhaustive investigation was not within the scope of this work; therefore , we identified a subset of candidate genes for experimental follow-up . Candidates were chosen by considering the significance of periodic expression and the t-statistic derived in the 2-population model . Additionally , we considered genes that had plausible , but not well defined connections to hair growth and regulation as determined by the literature . We chose Signal Transducer And Activator Of Transcription 5A ( Stat5a ) , Fermitin Family Member 2 ( Fermt2 ) and Vimentin ( Vim ) for candidates associated to the background population , as well as Ovo-Like 1 ( Ovol1 ) and SMAD Family Member 6 ( Smad6 ) for the expanding population . In the follicle dissection study of Rendl et al . [36] , Stat5a and Ovol1 were also identified as signature genes for DP and MX cells , respectively; however , the other candidates were not linked to a specific cell type in the same study . The metrics relating to these candidate genes as well as others identified in this study are presented in Supplementary File S2 . For candidate and control genes , we confirmed the expected relative expression profiles by qRT-PCR ( Supplementary Figure S8B , C ) . Specifically , we confirmed that the relative expression of background candidates decreased during anagen onset and then increased after completion of anagen with maximums observed in or near the telogen phase , while the expanding population demonstrated a reciprocal profile ( recall Figure 1 , cluster 2 and 1 respectively ) . We investigated the localization of candidate gene products by various imaging techniques in samples corresponding to telogen taken before induction ( day 0 ) and anagen taken 16 days after induction ( day 16 ) . Here , we recall that background candidates were determined to be markers for hair follicle cell populations that remain relatively stable throughout the hair cycle; these were also enriched with DP signature genes . Using ISH and fibroblast growth factor 7 ( Fgf7 ) as a control marker [42] , we confirmed that the signature gene Stat5a , also identified by the 2-population model , was expressed in DP cells ( Figure 6A R1 ) . We note that technical negative and positive controls for ISH can be seen in Supplementary Figure S9 . Interestingly , ISH imaging produced similar expression results for background candidates in both telogen ( day 0 ) and anagen phases ( day 16 ) while qRT-PCR and microarray suggested notable differences in relative expression between these phases . These observations were explained by the 2-population model above , where increases in the expanding population decreases the relative size of the background population , resulting in notable expression changes for mixed cell population samples , such as the whole skin samples used in qRT-PCR . Although consistent with our expectations as described , for completeness we add that insensitivity in ISH imaging could provide another explanation for the observations . Furthermore , significant changes of Fgf7 expression have been detected in DP isolates [37] , although under different conditions from the data considered here . We observed the same localization pattern in the novel candidate marker , Fermt2 . Roles for Fermt2 have been proposed in regulation of the extra-cellular matrix , actin organization as well as cell-ECM focal adhesions [43] . It is also associated with β-catenin/TCF4 complex , and knockdown of Fermt2 leads to loss of β-catenin mediated transcription [44] , ultimately affecting myogenic development . Other effectors of the β-catenin/TCF4 complex , such as Wnt , are known to be required for the hair inducing property of DP cells [45] . Background population proteins were also localized by immunofluorescence . The morphology , as determined by brightfield and DAPI staining , was used to identify DP localization . We observed localization of Stat5a protein to the DP in anagen samples; however , localization was much more difficult to assess in telogen samples ( Figure 6B R1 ) . Technical issues prevented antibody staining for Fermt2; therefore , we considered an alternate candidate , Vim , which was also associated to the background population . Immunofluorescence demonstrated Vim protein expression localized to the cytoplasm of DP cells in both telogen and anagen phases . We also observed Vim staining in the dermal sheath that surrounds the anagen bulb ( Figure 6B R2 ) . While Vim expression in the follicle has been reported [36] , [46] , [47] , it is expressed in other dermal cells and it is thus not specific to the hair follicle . Its expression in both DP cells and in cells adjacent to the hair follicle , or macro-environment , emphasizes the importance of the use of whole skin in our study . For example , Plikus et al . [48] , [49] demonstrated that the macro-environment can be the source of paracrine signals that influence the hair cycle . Imaging results also confirmed candidate markers for the expanding population . Here we recall that the expanding population candidates were determined to be markers for cells whose relative population size increases during anagen , followed by a sharp decline in catagen ( Figure 4 ) ; these were also enriched for MX cell genes . Using Forkhead box protein N1 ( Foxn1 ) as a positive control for MX cells [50] , note: Foxn1 was also identified in the current study as a candidate marker for the expanding population ) , we observed mRNA expression of the signature gene Ovol1 in the proliferating cell populations of the hair shaft in anagen samples , as described in the literature [36] , [50]–[52] , Figure 7 R1 ) . We also observed the same expression pattern and localization to the hair shaft for the novel candidate marker Smad6 ( Figure 7A R2 , day 16 ) . Further evidence of an association between these candidate markers and the expanding population was demonstrated by a general lack of staining in telogen samples . The 2-population model attributes this lack of expression to the absence of the expanding cells in the telogen phase ( Figure 4 ) . Additionally , immunofluorescence confirmed the localization of Ovol1 and Smad6 protein near the MX cell marker Foxn1 ( Figure 7B ) . Matching the pattern of Foxn1 expression at day 16 , ovol1 and Smad6 stained the epithelial , matrix-derived inner root sheath cells and the upper part of the matrix cells that surround the hair bulb while Smad6 was also detected in the epithelial outer root sheath . Although their general expression was observed , Vimentin and Smad6 were not identified as signature genes for DP and matrix cells respectively by Rendl et al [36] . The results here ( Figures 6 and 7 ) demonstrate their expression in distinct compartments relating to follicle epithelial ( Smad6 ) and background mesenchymal ( Vim ) cells . The staining results and the 2-population model support the hypothesis that the computationally identified expanding population was associated to follicle epithelial cells . Currently there is no direct evidence of Smad6 in hair cycle regulation; however , Smad6 is a well-known negative regulator of the Tgfβ signaling pathway [53] , [54] . Given its role in regulating Tgfβ signaling , as well as its proximity to MX cell markers , Smad6 may be an important candidate for future study in hair cycle regulation . We also noted the periodic expression of bone morphogenetic protein ( BMP ) genes , which have been documented as important regulators of skin and hair development . Four BMP genes showed periodic expression as LFOs: BMP 1 was found in cluster 1 ( dermal papilla-associated ) while BMPs 2K , 8a , and 7 were expressed in cluster 2 ( matrix-associated ) . Other BMPs such as 2 and 4 were present in the original data , but the expression data contained too much variability to survive the FDR cutoff . The complete lists of LFO genes that matched the background and expanding population clusters , along with statistical metrics , is in Supplementary File S3 . The coupled oscillator model suggests that out-of-phase clustering is maintained by positive and negative coupling . The two population model indicates that these clusters are associated to specific cell populations . Taken together this is similar to negative feedback , for example the expanding population may drive the background population to produce an inhibitory signal , such as apoptosis , that in turn depletes the expanding population . However , if the background population is static , how is it contributing to such a control loop ? For example , when the expanding population is relatively high , one would expect an increase in the expression of the inhibitory genes from the background to drive down the expanding population . One reasonable possibility is that expression changes were occurring within the background population . On average we found that the assumption of static intracellular expression was reasonable enough to estimate population dynamics ( recall Figure 4 ) ; however , many individual expression profiles were poorly described by this assumption ( recall Figure 5A ) . It is possible that these poorly described genes were undergoing intracellular changes , and could be responsible for the physical communication of inhibitory signals from the static background to the expanding population . We investigated the possibility that inhibitory signaling genes may be in the DP enriched group identified as LFO cluster 2 , but not well described by the static intracellular expression model . We expect such signaling genes to display an increased expression 14 to 16 days after morphogenesis , near the on-set of catagen and before the sharp decline in the expanding population ( Figure 4 ) . Using this criterion , we identified 88 expression signals ( relating to 74 unique genes; see Supplementary File S4 ) . We observed that these expression signals , on average , are consistent with population driven changes until near catagen on-set , where they begin to increase more than what was explained by static intracellular expression assumed in the 2-population model ( Supplementary Figure S10 ) . Of these genes , 50 were annotated as extracellular genes which yields and enrichment p-value of 7 . 46E-18 , improved enrichment over cluster 2 with a p-value of 3 . 63E-8 . For a full list of significant enrichment categories see Supplementary File S5 . Interestingly , this relatively short list includes Tgfβ2 , which is currently thought to be one of the signaling molecules produced in DP cells to initiate apoptosis in hair epithelial cells at catagen on-set [7] . Given the observed expression signal , membership in DP enriched cluster 2 , high enrichment for extracellular genes and inclusion of Tgfβ2 , this list may contain potential targets for molecules that communicate an inhibitory signal from the DP to proliferating hair epithelial cells , closing a negative feedback loop . Obviously further experiments will be required to test this hypothesis; however , it does provide a starting point for future validation of the conclusions drawn above and , perhaps , even those identified in the model of Al-Nuaimi et al . [10] . Although our approach provides novel insights and genes associated to the hair follicle , we also recognize that there are several limitations to this study . We studied microarray-derived RNA expression data from developing mouse skin that included non-periodic as well as periodic gene expression patterns . Due to the cyclic nature of the hair cycle , we chose to focus our study on the latter . We emphasize that our study would overlook important regulators of the hair cycle if they were not periodically expressed . Next , we only considered a single time course , experimental study ( Lin et al , 2009 ) , which obviously limits the data and conditions available to us ( sparse sampling , limited technical replicate measurements and inclusion of only early cycles ) , and could lead to some important genes and cycle dynamics being excluded from further analysis . Furthermore , biological and technical variation , along with typical tradeoffs in sensitivity versus specificity associated with parameter selection , such as p-value thresholds , will further limit statistical detection of important mRNAs or expression patterns . Due to concerns of batch effects , we did not choose to combine additional datasets from other experimental studies to offset these issues . Instead , we chose to limit the scope of our investigation to describing a specific , but prevalent , dynamic pattern observed in the data . Again , by limiting the scope in this manner , it is likely that some important hair cycle regulators were overlooked . For example , BMP 2 and 4 have been shown to influence anagen initiation [48] , [49] , but due to sample variability these patterns fell outside the range of detection for this study . However , our investigation did encompass over 3 , 000 unique genes , where follow-up dynamic , enrichment and experimental analyses all suggested a possible role in the hair follicle and cycle dynamics . Study design also limited us to time course over a single cycle of follicle synchronized hair growth . We were not able to test if the identified expression patterns , specifically synchronized out-of-phase gene expression , continued for additional cycles . This is a typical experimental limitation due to the loss of follicle synchronization as animals mature , at latter stages of hair growth . This is a different concept from the synchronization describing gene expression patterns . One might still expect that similar gene expression patterns within individual follicles , and the surrounding microenvironment , continue with additional cycles; however , without single follicle tracking , we cannot confirm this . Furthermore , our dynamic coupled oscillator model would never predict such follicle-level de-synchronization , as we did not include any mechanisms for cycle variability nor did we include the concept of individual follicles . Accounting for stochastic variation and spatially modeling individual follicles that are themselves coupled , represents an additional level of complexity that may more accurately model the rich dynamics of the hair system , but was not considered in this study . Finally , we modeled gene expression from whole skin , since isolation of hair follicles prior to gene expression profiling is resource intensive and was beyond the scope of our work . In doing so , we relied on the 2-population model , cell type specific enrichment ( based on experimentally purified cell populations [36] , [37] ) and experimental imaging to make associations between computationally derived gene groupings and distinct biological populations . While these results were very encouraging , we would like to emphasizes that the computationally derived populations do not represent a specific , predefined cell type . Supplementary Figure S7A shows that the majority of the signature genes were not identified , and the populations contained signature genes from multiple cell types . However , we do note that even experimentally derived gene signatures also demonstrate overlap ( see Supplementary Figure S7B ) . Furthermore , it likely that cell types not investigated by enrichment also contributed to the estimated populations , such as endothelial cells involved in capillary network remodeling , adipocytes or immune cells that may have active roles in hair growth . Ultimately , we identified many associations between the computationally derived gene groupings and distinct , hair cycle relevant cell populations , but we cannot exclude that gene expression unrelated to the hair follicle or hair growth may have contributed to both false positives and negatives . In this study , we focused on potentially periodic gene expression patterns in whole skin that changed on the same time-scale as cyclical hair growth . We identified two distinct clusters consistent with synchronized , out-of-phase gene expression ( Figure 1 and 2 ) . Through nonlinear-dynamic analysis , we proved that a simple , coupled oscillator model was mathematically sufficient to recapitulate this observed synchronization , and that the coupling scheme involves both positive and negative coupling ( Figure 3A ) . We go on to show that these clusters can be associated with either static or expanding cell populations ( Figures 4 and 5 ) , and that the size of the expanding population , determined by gene expression data , was consistent with the population dynamics of follicle epithelial cells ( Figure 4 ) . Follow-up experimental and enrichment analyses indicated that the corresponding genes ( provided as Supplementary Information File S3 ) were strongly associated with biologically distinct cell-types , such as MX or DP cells ( Figures 6 and 7 , Supplementary Table S1 ) . Taken together , these results were consistent with regulatory mechanisms involving negative feedback from background mesenchymal cells to the expanding epithelial cells ( see summary Fig 8 ) . Finally , we identified a subset of genes that could potentially communicate the inhibitory signal to the follicle ( provided as Supplementary Information File S5 ) . Other aspects of the study provided interesting , but speculative , insights on possible alternative hair cycle states that are similar to those of miniaturized hair follicles ( Figure 3B ) . The model of hair cycle presented here suggests some role for proliferating follicle epithelial cells to be regulated by a systems-level inhibitory response , likely to emanate from the DP . Conceptually , regulated apoptosis could be one way in which a large number of genes from the same general population of cells are inhibited by a second population . This mechanism has also been explored in a kinetic model of hair cycle which shows that negative feedback via DP regulated apoptosis is sufficient to account for the cyclical nature of hair growth [10] . Because mRNA expression data underlies the model presented here , we were able to advance this idea a step further and identify candidate signaling proteins based on the dynamics of the gene expression . Although experimental confirmation of these candidates was beyond the scope of the investigation , we note that the list was highly enriched for extracellular proteins and with only 74 genes we were still able to identify Tgfβ2 as a possible candidate . For clarity , we emphasize here that negative coupling is not the only aspect of our model , which also includes intrinsic oscillations and positive coupling . In fact , our model predicts that reduction of the DP associated oscillators actually results in a shortened cycle ( see Figure 3B ) . Furthermore , the true physical mechanisms underlying the hair cycle are likely far more complex , and studies show that anagen length and hair size actually decrease upon depletion of DP cells [55] . Both our computational and existing experimental results suggest that inhibitory regulation of MX derived cells cannot describe all aspects of the hair cycle; however , it is likely to play an important role , with one possibility being regulated apoptosis [7] , [56] , and the genes we have identified here could help guide follow-up experiments . Finally , the most intriguing aspect of this study was the predicted proximity of the hair system to a critical phase transition ( Figure 3B ) . In the observed dynamics the hair system was in a stable -state , in which the proportion of positive and negative oscillators may vary , or at least increase , without affecting the overall period . However , a decrease in this proportion , for example a reduction in positively coupled oscillators , would throw the system in to a travelling wave state , where the two clusters drive each other into higher frequency oscillations . If this behaviour can be validated , it would have important biological implications . Biologically the model suggests that there is a systems-level regulation within and between genes related to follicle epithelial and background mesenchymal cell populations , which represents a balance of negative and positive driving forces . Loss of regulation of the DP population , for example , would mean a decrease in negative forces acting on the epithelial population and would throw the system into a fundamentally different state . This new state would be characterized by a drastically reduced period of expression , similar to hair miniaturization resulting from androgenetic alopecia . Unfortunately , testing of this hypothesis could prove difficult . Removal of several genes via genetic knockout would have consequences not accounted for here; however , inhibition of the physical signaling between the DP and MX derived populations may be more tractable . Excitingly , a recent study by Rompolas et al . [57] introduced new methodology to explore such interactions of the hair follicle in live mice using laser ablation of DP cells . The approach not only allowed for the elimination of a specific cell population , but also removed technical complications associated with follicle synchronization , as individual follicles could be monitored over time . It would be of future interest to see if this methodology could be modified to properly test the predictions presented herein . To build on the insights of the present framework , we can offer several additional directions for future work . For example , there are several possible advancements to the coupled oscillator model , including: analytically solving the existing coupling scheme for excitable elements ( similar to [58] ) opposed to oscillators would better capture the pulsing behavior of gene expression and hair growth; integrating noise , known to have a major impact of synchronization [59] , could help capture both expression and cycle variability; and coupling together multiple coupled systems could capture associations and variation between follicles . Experimentally , new time course data could identify new behaviors . Performing a single extensive time course from morphology to end of second round of hair growth would capture anticipated transients and determine a proper time scale for synchronization . Increasing the sampling frequency could identify high frequency oscillators and perhaps provide a means to couple circadian rhythm to the current system . Given that we systematically identified two clusters from whole skin data , a beneficial advancement would be to directly collect data from follicle specific cell types , such as MX and DP cells . Producing a time course using purification techniques similar to Rendl et al . would be the most direct way to prevent confounding expression signals from non-relevant cell types and provide a specific interpretation of modeled populations . In our experience , any additional time courses would need to have , at a minimum , a sampling rate 3 times that of the desired frequency for identification . However , we would strongly suggest doubling the number of points in the time course and including 3 replicates at each point for statistical and modeling considerations . Our hope is , that by incremental advancement , the framework provided here can be used to bridge the gap between high-throughput measurement data and systems-level properties of hair cycling .
This study was performed in strict accordance US Animal Welfare Regulations at an AAALAC accredited site . The research protocol was approved by the Institutional Animal Care and Use Committee of Procter & Gamble . Every effort was made to minimize suffering of all animal subjects . We employed data originally generated in Lin et al . [11] . The authors profiled both second , naturally synchronized and depletion-induced hair cycles . Samples were collected from the upper-mid region of C57Bl/6 mice and analyzed using Affymetrix Mouse Genome 430 2 . 0 . For additional experimental details please refer to the original article . The raw intensity data was collected from the NCBI Gene Expression Omnibus as accession number GSE11186 . The data was uploaded in CEL file format and preprocessed for quality control . Sample GSM281802 was removed based on suspected RNA degradation , a mean correlation coefficient less than 0 . 95 , multiple outliers determined by the MA plot , and high background error and variation determined by RMA . The remaining samples were summarized and normalized using the RMA function from the Bioconductor ‘affy’ library in R , applying quantile normalization and RMA background correction from affy version 1 . 1 . A log base 2 transform was applied to the expression data for all subsequent analysis except for the 2-population mixture model . An R script containing the general QC and RMA methods used can be found in the Supplementary File S6 . The two experimental conditions corresponding to the natural and induced hair cycles , were combined into a single time course as suggested by the original authors . The five sampled time points for the induced cycle , {3 , 5 , 8 , 12 , 17} days , were mapped to time points in the natural cycle , {24 , 25 , 27 , 29 , 37} days , based on the morphology of skin sections . Hair morphogenesis during synchronous growth was established by histologic criteria [11] . Combining samples with similar morphologies , and therefore in similar hair cycle phases , we expect to limit the variability of gene expression that is associated to the hair cycle phenotype . We visualized the expression data using standard heat maps . Multiple values at a given time point were averaged . For visualizing expression values of different scales in a single image it was necessary to normalize the data . Two different normalization schemes were used . Figure 1 focuses on relative levels of periodically expressed genes , here we used a 0–1 normalization: . The associated time scale is ‘days postnatal’ and corresponds to the natural second cycle; the induced cycle was matched to time points as discussed above . In Figure 5 and the 2-population mixture model , we applied a fraction of max normalization , , to capture information of relative fold change in expression . The associated time scale was days from initiation to emphasize differences in the natural and induced cycle , which is assumed to begin after morphogenesis , which is ≈23 days , and depletion , respectively . The normalization schemes described here were for visualization purposes , and were not used in any statistical analysis . For all time series , we present a color bar to roughly indicate the corresponding phase of the hair cycle , the timing for the color bar was taken from [11] . We employed a mathematical description of coupled oscillators to study general features of the synchronization observed in the hair cycle data . We considered a modified version of the simple Kuramoto model [16] suggested by Hong and Strogatz [33] . Hong and Strogatz show that a two group model , one positive and one negative , is sufficient to spontaneously produce two out-of-phase clusters . Oscillators which are positively or negatively coupled will be drawn towards or pushed away from other oscillators on the unit circle , respectively . After some simplification and incorporation of EQ 9 the governing equations reduce to ( 10 ) where is the phase of the th oscillator , calculated by EQ 7 , which is assigned to group , is the natural or intrinsic frequency for , is the coupling constant for group , and from EQ 9 where the subscript is dropped for simplicity . The dot denotes change with respect to time , and can be calculated using EQ 8 . Recall that oscillators represent probesets with expression patterns identified as periodic . Here we assume two groups where and , , . Introducing these two groups to EQ 10 we have ( 11 ) where is the number of positively coupled oscillators . We note that EQ 11 is identical to EQ 2 presented in the Results and Discussion , and was included here only to maintain the continuity of the method descriptions . We also reemphasize several simplifications in this formulation . First , the model is a mean field approximation in which each individual oscillator is connected to all other oscillators through the order parameter , . This is derived from an assumed all to all connectivity , which is obviously not expected in a gene network; however , the mean field approximation works well if the effective coupling on the oscillators ( or genes ) is well described by an average of the individual couplings . Such models have successfully described high level properties of many large , complex systems including statistical mechanics ( overview of several models [17] ) , economics [18] , [19] and even social networks [20] . Second , we note that the variables here are considered independent , for example the model assumes that the proportion of oscillators can be varied without affecting other independent variables , such as the ‘intrinsic’ oscillations , . However , in reality removal of genes from the system will have an impact not captured in the model , such as an alteration or even cessation of the assumed ‘intrinsic’ oscillations . Finally , we emphasize that the coupling describes oscillator interactions and not necessarily the underlying driving force for oscillation , which is typically attributed to . With this level of abstraction and simplification it was not possible to describe most of the details of the hair cycle including mechanistic molecular connectivity; however , we were able to describe more general aspects of the system such as a stable , synchronized state . To solve the system , we follow the original paper [33] , and summarize the process here for the reader's convenience . We can reduce the model to a low dimensional system in terms of the first order parameters for each group . First , we consider a system where , we validate the use of this assumption later . Second , we assume the were randomly distributed via a Lorentzian probability distribution . Importantly , we note that use of a single frequency , opposed to a distribution , will not recapitulate the distributed phases observed in Figure 1C [67] . Here , we have moved to a rotating frame such that the mean of is zero; in our system this was done by subtracting the principal periodic component , radians per day . Finally , we can apply the ansatz of Ott and Antonsen [68] which yields ( 12 ) where and are the first order parameters ( similar to EQ 9 with ) for the two oscillator groups related to positive and negative coupling , respectively; is the proportion of positively coupled oscillators and . The bar denotes complex conjugate . We note that EQ 12 is identical to EQ 3 presented in the Results and Discussion , and was included here only to maintain the continuity of the method descriptions . Using EQ 12 we can solve for the critical value of such that only the incoherent state is stable when and is , therefore , the lower bound for observing synchronization . We can also estimate other bifurcation points of the system , and . For more details see [33] . We solved for various properties of the hair cycle system using EQ 12 and the oscillator state variables solved for above . We assumed the observed period of the hair cycle system was at a quasi-steady-state , where the magnitude of the first order parameter is constant and the rate of change of the phase is also constant . This was demonstrated by observed in Figure 2A . Given the quasi-stead-state , we solved EQ 12 in a rotating frame described above , allowing us to set the left hand side to zero . We considered two possible configurations of assigning clusters ( from Figure 1C ) to positive or negative coupling groups . After coupling assignment , we calculated , , and from the data and solved EQ 12 for the unknown parameters and . Then and were used to solve for . We note that EQ 12 was solved by letting and lie on the real axis so they were equivalent to and , respectively . This assignment can be done , without loss of generality , for a quasi-steady-state , out-of-phase system . For the configuration with ( clust1 ) = ( + ) and ( clust2 ) = ( − ) , we found that the system was unstable ( Figure 3A purple dashed line ) . We calculated and which is not physically realizable . However , we found the opposite configuration , with ( clust1 ) = ( − ) and ( clust2 ) = ( + ) , to be a stable solution with radians per day and ( Figure 3A red solid line , recall actual data in Figure 2A red , see Supplementary File S2 for a simulation describing individual oscillators with these properties ) . The bifurcation diagram was solved numerically ( Figure 2B ) . We found the long time solutions to the system of ordinary differential equation ( EQ 12 ) for various values of while holding all other variables constant . We found to be sufficient . We verified the assumption of by simulating the low dimensional system , EQ 12 , and comparing that to simulations of the high dimensional , EQ 10 , with ( Supplementary Figure S13 ) . Given noise , due to initial configurations , associated with the finite , high dimensional system , 100 iterations were calculated and the mean and standard deviation were reported . The two representations are sufficiently similar and have nearly identical steady-states . All numerical simulations were performed in Matlab [69] using ‘ode45’ . All Matlab scripts necessary to solve for model variables and reproduce simulations found in figures and movies are available in the folder ‘meanField’ in Supplementary File S6 . Observations of two distinct gene expression clusters motivated us to explore possible relationships to different cell populations within the hair follicle . We considered the scenario in which observed expression changes are due to changes in relative cell population size as opposed to intracellular changes . In Supplementary Figure S3 , we showed a simple example in which and had a high and low concentration in pop 1 , respectively , and reciprocal concentrations in pop 2 . While varying the size of pop 1 , holding the pop 2 size constant and holding all internal concentration level constant , we can see that the observed concentration of and change , and do so in an out-of-phase manner . To explore this in our system we wished to reverse the process and estimate the relative sizes and intracellular expressions given the observed mixed expression . To achieve this we applied in silico microdissection [35] . Briefly , in silico microdissection works by applying a simple linear model of mixed samples ( 13 ) where is the observed expression of gene in mixed sample ; indicates the intracellular expression in populations and ; and is the cell fraction of in mixed sample . Given the cell fraction , , for each sample we can solve EQ 13 for the internal concentration in the two populations . In this situation each gene is an independent problem , each solved via simple linear regression over all mixed samples . This is an overdetermined system if the number of populations considered is less then the number of samples . We can also recast the problem to solve for given the internal concentration for each population . In this situation each mixed sample is now an independent problem , each a constrained linear problem over all genes . Here is constrained between 0 and 1 , the problem is convex and can be solved efficiently . An iterative process , similar to expectation-maximization , can be used to solve for both and the y's simultaneously . We consider a model of an expanding cell population mixed with a constant background population . We treated the hair cycle expression chips as independent mixed samples each with possibly different cell fractions . No information of cycle type or time was needed , nor any strategy for combining samples as in the previous periodic identification . For later comparisons of the induced and natural cycle , we set the time relative to cycle initiation , which we assumed to be after morphogenesis ( postnatal day 23 ) or after depletion . This time scale was only used for graphical representation , and was not used in any calculations . A linearly increasing function from 0 to 1 was used as the initial conditions for , the cell fraction of the expanding population . We also included the expression of all 45k+ probesets without the log2 transform , as suggested in [35] , all other preprocessing was the same . Given this , over determined , model we implemented the above iteration strategy to solve for and the relative size ( Figure 4 ) and the internal expression ( Figure 5C ) . We note here that it is not reasonable to expect all intracellular expression to remain constant over the whole hair cycle; however , if the relative population change is large , as seen here , compared to the intracellular expression change for many genes then it is a reasonable assumption . While calculating the internal expression for the two populations , we also estimated the corresponding standard error using common methods associated with linear regression . The standard error was used to produce a t-statistic and p-value for each probeset , which indicated the extent to which a gene was differently expressed between the two populations ( Figure 5B ) . The probesets for the LFO subgroup with a t-statistic above a 0 . 10 false discovery rate were assigned to the population in which they were predicted to have higher expression . We note that nearly all of the LFOs met the statistical requirements , 3975 out of 3988 . This was equivalent to separating the probesets into two groups based on the estimated t-statistic , and was found to be equivalent to separation of LFOs by phase ( Supplementary Figure S6 ) . We also considered a computational negative control . In the above analysis , we inherently assumed that expression is related to time , after morphogenesis or after depletion . Our population analysis allowed us to then associate expression to relative population size , and therefore , plot relative population size as a function of time . Here we considered a negative control , that expression is random with respect to time , and not related to hair cycle . To achieve this , we randomly permuted ( or shuffled ) the time courses for each probeset . For a proper comparison the depletion and naturally induced time courses were not intermixed , and kept separate . After permutation , we employed the exact same analysis and plotting procedure as above ( used to produce Fig . 4 and 5 ) . The results are shown in Supplementary Figures S4 and S5 . In our negative control , we could see that there was no indication of expansion and depletion corresponding with anagen and catagen , respectively ( Supplementary Figure S4 ) . Furthermore , we did not observe improved statistics , as in coefficient of determination or the t-statistic ( Supplementary Figure S5 ) . We considered this sufficient evidence that the negative control produced only random population changes with respect to time , as expected . All code for estimating the two populations was written in Matlab and implemented by the scheme discussed in [35] . The constraint linear problem was solved using Matlab Optimization Toolbox function ‘lsqlin’ with default options . Standard errors were estimated using the Matlab Statistics Toolbox function ‘regstats’ . Matlab scripts for running all analysis described above and for generating the data in associated figures are available in the ‘2pop’ folder in Supplementary File S6 . We used the online tool DAVID 6 . 7 to perform basic enrichment analysis [70] . We used the full mouse genome as the background gene set . For biological process enrichment , we used the Gene Ontology annotations under ‘GOTERM_BP_FAT’ and for pathway enrichment , we used KEGG annotations under ‘KEGG_PATHWAY’ . The enrichment was done using different target sets indicated in the main text . We used the Normalized Google Distance ( NGD ) to estimate enrichment of genes related to hair . The NGD is a semantic similarity measure [71] , which for two terms and is defined as ( 14 ) where and are the numbers of pages the terms and are found in , respectively , is the number of co-occurrences and is the total number of pages considered . We note that the more frequent and co-occur the lower NGD will be , and that if then NGD = inf . Here we applied the search to all abstracts in the NCBI PubMed database . We calculated the NGD between the term ‘hair’ and all mouse gene symbols . All genes with an NGD to ‘hair’ of less than 1 . 0 were used as the final set of Hair related genes . We applied a standard enrichment test using a hypergeometric distribution . The target set was the list of all periodic genes and the background set was the full mouse genome . The threshold of 1 . 0 was chosen as it is the NGD of the expected value for independent or unrelated terms . Briefly , given a set number of occurrences for a term then the probability of finding term in pages is . Assuming that two terms , and are independent , we have . Plugging these values into EQ 14 we find that NGD = 1 . A cell type enrichment analysis was used to link model populations to specific cell types . Two existing studies , Rendl et al . [36] and Greco et al . [37] , dissected hair follicles into specific , predefined cell types relating the the hair follicle: dermal papilla , melanocytes , matrix cells and outer root sheath cells from [36] and Bulge cells from [37] . Using mRNA microarray data the studies defined gene signatures for each population as sets of probesets and corresponding genes with expression nearly exclusive to a particular cell type . Using these signatures to annotate probesets with a particular cell type , we applied standard enrichment tests using a hypergeometric distribution . The target set was the list of genes determined to be in the expanding or background population ( seen in Figure 5C ) and the background set was the full mouse genome . Male mice , C57Bl/6 ( Charles River Laboratories , Portage , MI ) at 62-66 days of age , in the telogen phase of the hair cycle [5] are shaved in the dorsal area ( area of 1 . 5 inches×2 inches ) followed by treatment with Nair ( Church & Dwight Co . ) to the same area for 1 hour before washing off to initiate the hair cycle . Nair depletion induces a similar response as wax by damaging the hair shaft to start a homogenous re-entry into anagen [72] . Mice were collected at various timepoints after induction treatment . ISH was performed using QuantiGene ViewRNA protocols ( Affymetrix , Santa Clara , CA ) . Five µm formalin fixed paraffin embedded ( FFPE ) sections were cut , fixed in 10% formaldehyde overnight at room temperature ( RT ) and digested with proteinase K ( Affymetrix , Santa Clara , CA ) . Sections were hybridized for 3 hours at 40°C with custom designed QuantiGene ViewRNA probes against specific target genes and the positive control genes used were Fgf7 for dermal papilla cells and Foxn1 for matrix cells ( Affymetrix , Santa Clara , CA ) . Bound probes were then amplified per protocol from Affymetrix using PreAmp and Amp molecules . Multiple Label Probe oligonucleotides conjugated to alkaline phosphatase ( LP-AP Type 1 ) were then added and Fast Red Substrate was used to produce signal ( red dots , Cy3 fluorescence ) . For two color assays , an LP-AP type 6 probe was used with Fast Blue substrate ( blue dots , Cy5 fluorescence ) followed by LP-AP type 1 probe with Fast Red Substrate ( red dots , Cy3 fluorescence ) to produce a dual colorimetric and fluorescent signal . The probes sets used for ISH are described in Table S2 . Slides were then counterstained with hematoxylin . Serial sections were also subjected to hematoxylin and eosin staining per standard methods to confirm the identity of cells in the region of ISH signals . Images were collected using a Deltavision microscope ( Applied Precision ) , and the fluorescent images were created using softWoRx 5 . 0 ( Applied Precision ) . The in situ hybridization assay in this study utilizes branched DNA ( bDNA ) technology , which offers near single copy mRNA sensitivity in individual cells . The bDNA assay uses a sandwich-based hybridization method which relies on bDNA molecules to amplify the signal from target mRNA molecules . Each probe set hybridizing to a single target contains 20 oligonucleotides pairs . This was followed by sequential hybridization with the final conjugation of a fluorescent dye . Thus , each fully assembled signal amplification ‘tree’ has 400 binding sites for each labeled probe . Finally , when all target specific oligonucleotides in the probe set have bound to the target mRNA transcript , the resulting amplification of signal approaches 8000-fold ( 20 oligonucleotides times 400 binding sites = 8000 fold ) . Immunofluorescence staining was performed on fresh frozen cryosections ( 10 µM thickness ) or FFPE sections ( 5 µM thickness ) of mouse skin to visualize the hair follicles present at Day 0 and Day 16 . Cryosections were stored at −80°C until use . Cryosections were dried for 30 min at room temperature and fixed by immersion in ice-cold acetone for 10 mins . Cryosections were then air-dried for 5 mins and washed three times with PBS . For FFPE sections , deparaffinzation was performed using xylene and series of alcohol changes . Antigen retrieval for performed using 0 . 05% trypsin at 37°C for 20 mins . Both cryosections and FFPE sections underwent the same treatment after this step . The sections were blocked for 1 hour using normal donkey serum ( NDS , dilution 1∶10; Sigma-Adhrich ) in PBS . Sections were then incubated with specific primary antibodies ( as described in Table S3 ) in 1∶5 dilution of blocking solution for overnight at 4°C and then washed three times with PBS . Next , sections were incubated with Alexafluor-488-conjugated donkey anti-rabbit and Alexafluor-594-conjugated donkey anti-goat antibodies ( Vector Laboratories , 1∶500 ) for 1h at 37°C , washed three times with PBS . Final wash was performed with DAPI and the sections were mounted using Flurosav ( Calbiochem ) . Images were collected using a Deltavision microscope ( Applied Precision ) , and the fluorescent images were created using softWoRx 5 . 0 ( Applied Precision ) . Foxn1 was used as a positive control for matrix cells based on previous literature [50] . Morphology , as determined by brightfield and DAPI staining , was used to identify DP localization . We considered Fgf7 [42] as a positive control for DP localization; however , all antibodies tested showed significant non-specific staining . Total RNA was extracted from mouse skin samples at days 6 , 16 , 23 , 29 , 38 , 44 and 59 using Agilent's Total RNA isolation mini kit ( Agilent Technologies ) . Reverse transcription reaction was performed with 500 ng of total RNA using the Superscript VILO cDNA synthesis kit ( Life technologies ) . A 1∶25 dilution of cDNA was used in the QRT PCR reaction . QRT-PCR was carried out in a 10 µl reaction mixture with gene-specific primers and β-Actin using RT2 SYBR Green ROX qPCR Mastermix ( Qiagen ) . The PCR conditions were 95°C for 10 min , and 40 cycles of 95°C for 15 s , 59°C for 30 s , 72°C for 30 s on the ABI HT 7600 PCR instrument . All samples were assayed in quadruplicate . The differences in expression of specific gene product were evaluated using a relative quantification method where the expression of specific gene was normalized to the level of β-Actin . Primer sequences available in Supplementary Table S4 . | The hair cycle represents a complex process of particular interest in the study of regulated proliferation , apoptosis and differentiation . While various modeling strategies are presented in the literature , none attempt to link extensive molecular details , provided by high-throughput experiments , with high-level , system properties . Thus , we re-analyzed a previously published mRNA expression time course study and found that we could readily identify a sizeable subset of genes that was expressed in synchrony with the hair cycle itself . The data is summarized in a dynamic , mathematical model of coupled oscillators . We demonstrate that a particular coupling scheme is sufficient to explain the observed synchronization . Further analysis associated specific expression patterns to general yet distinct cell populations , background mesenchymal and rapidly expanding follicular epithelial cells . Experimental imaging results are presented to show the localization of candidate genes from each population . Taken together , the results describe a possible mechanism for regulation between epithelial and mesenchymal populations . We also described an alternate state similar to hair miniaturization , which is predicted by the oscillator model . This study exemplifies the strengths of combining systems-level analysis with high-throughput experimental data to obtain a novel view of a complex system such as the hair cycle . | [
"Abstract",
"Introduction",
"Results/Discussion",
"Methods"
] | [
"animal",
"models",
"systems",
"biology",
"computer",
"and",
"information",
"sciences",
"model",
"organisms",
"systems",
"science",
"mathematics",
"biology",
"and",
"life",
"sciences",
"mouse",
"models",
"physical",
"sciences",
"nonlinear",
"dynamics",
"research",
"and",
"analysis",
"methods"
] | 2014 | Mouse Hair Cycle Expression Dynamics Modeled as Coupled Mesenchymal and Epithelial Oscillators |
The human nuclear poly ( A ) -binding protein PABPN1 has been implicated in the decay of nuclear noncoding RNAs ( ncRNAs ) . In addition , PABPN1 promotes hyperadenylation by stimulating poly ( A ) -polymerases ( PAPα/γ ) , but this activity has not previously been linked to the decay of endogenous transcripts . Moreover , the mechanisms underlying target specificity have remained elusive . Here , we inactivated PAP-dependent hyperadenylation in cells by two independent mechanisms and used an RNA-seq approach to identify endogenous targets . We observed the upregulation of various ncRNAs , including snoRNA host genes , primary miRNA transcripts , and promoter upstream antisense RNAs , confirming that hyperadenylation is broadly required for the degradation of PABPN1-targets . In addition , we found that mRNAs with retained introns are susceptible to PABPN1 and PAPα/γ-mediated decay ( PPD ) . Transcripts are targeted for degradation due to inefficient export , which is a consequence of reduced intron number or incomplete splicing . Additional investigation showed that a genetically-encoded poly ( A ) tail is sufficient to drive decay , suggesting that degradation occurs independently of the canonical cleavage and polyadenylation reaction . Surprisingly , treatment with transcription inhibitors uncouples polyadenylation from decay , leading to runaway hyperadenylation of nuclear decay targets . We conclude that PPD is an important mammalian nuclear RNA decay pathway for the removal of poorly spliced and nuclear-retained transcripts .
Eukaryotic messenger RNAs ( mRNAs ) undergo a series of maturation events before they are exported to the cytoplasm and translated . The complexity of alternative processing increases the likelihood of mistakes that produce aberrant mRNAs encoding defective proteins . In addition , pervasive transcription occurs across nearly the entire mammalian genome resulting in the generation of nonfunctional RNAs . Consequently , cells have evolved RNA quality control ( QC ) pathways to eliminate these RNAs [1 , 2] . The best-characterized RNA QC pathway is nonsense-mediated mRNA decay ( NMD ) [3] . NMD targets cytoplasmic mRNAs with premature termination codons ( PTCs ) , a potentially dangerous class of RNAs that produce truncated and possibly dominant-negative proteins . NMD is limited in at least three important ways . First , NMD recognizes PTC-containing transcripts upon translation , so each defective transcript still produces one polypeptide . This could be harmful to cells for highly transcribed NMD targets or particularly toxic polypeptides . Second , NMD is stimulated by the presence of a splice junction to identify PTCs , so transcripts from intronless genes will generally not be recognized . Third , pervasive transcription produces nuclear transcripts that would not be targeted by the cytoplasmic NMD machinery . Cells have additional nuclear RNA QC pathways to degrade RNAs not targeted by NMD , but the mechanisms involved remain unclear . Recently , functions for the nuclear poly ( A ) binding protein PABPN1 in RNA decay has been reported [4–6] . An RNA-seq study showed that knockdown of PABPN1 increases the accumulation of endogenous long noncoding RNAs ( lncRNAs ) , several noncoding snoRNA host genes ( ncSNHGs ) and transcripts upstream of mRNA gene promoters [4] . In addition , the Kaposi’s sarcoma-associated herpesvirus ( KSHV ) produces an abundant polyadenylated nuclear ( PAN ) RNA during the lytic phase of viral infection . A cis-acting element , called the ENE , protects PAN RNA from PABPN1-mediated decay by forming a triple helix with the poly ( A ) tail [5 , 7 , 8] . PABPN1 additionally promotes the degradation of a poorly exported intronless β-globin mRNA , but not its spliced and efficiently exported counterpart , suggesting it serves a QC function for non-exportable polyadenylated RNAs . PABPN1-mediated decay has been observed in S . pombe and humans suggesting an important conserved function [9–12] . The canonical mammalian poly ( A ) polymerases PAPα and PAPγ ( PAP ) , and the nuclear exosome are involved in PABPN1-mediated decay of intronless β-globin and PANΔENE reporters [5] . Several observations demonstrate that hyperadenylation by PAP promotes decay . First , knockdown of either PABPN1 or PAP stabilizes RNAs with shorter poly ( A ) tails . Second , knockdown of the exosome leads to the accumulation of hyperadenylated products . Third , inhibition of polyadenylation by cordycepin inhibits RNA decay . Fourth , expression of a dominant-negative PABPN1 double point mutant ( L119A/L136A or LALA ) that binds RNA but cannot stimulate PAP [13] stabilizes target RNAs . A global decay function for PAP is validated by the analyses reported here , so we now refer to this pathway as PABPN1 and PAPα/γ-mediated RNA decay ( PPD ) . PABPN1 and PAP have been extensively characterized for their roles in mRNA 3´-end formation [14] . Polyadenylation is initiated by co-transcriptional recruitment of the cleavage and polyadenylation specificity factor ( CPSF ) to the AAUAAA polyadenylation signal ( PAS ) through the CPSF30 and WDR33 subunits [15 , 16] . Extensive in vitro studies defined the roles of PAP , PABPN1 , and CPSF in the normal polyadenylation of mRNA 3´-ends [13 , 17] . Without CPSF , PAP has low binding affinity for RNA , but the CPSF-PAP interaction drives binding and generation of an oligo ( A ) tail . PABPN1 binds the oligo ( A ) tail and forms a complex with PAP-CPSF-oligo ( A ) . PAP becomes tightly tethered to the RNA , and polyadenylation is highly processive to ~200–300 nt poly ( A ) length . At this point , the interaction between PAP and CPSF is lost and polyadenylation becomes distributive , but this distributive polyadenylation continues to be stimulated by PABPN1 . We proposed that PABPN1-dependent and CPSF-independent stimulation of distributive PAP activity provides the polyadenylation associated with PPD [5] . Here , we refer to this as “hyperadenylation” as it occurs after the initial 3´-end formation step . To explore this globally , we performed RNA-seq following inactivation of hyperadenylation by two distinct methods . Consistent with the PABPN1 knockdown studies , we found that several classes of lncRNAs , including ncSNHGs , primary microRNA transcripts , and upstream antisense RNAs , are susceptible to PPD . In addition , we identified mRNAs and ( pre- ) mRNAs with retained introns that are PPD targets . Surprisingly , transcription inhibition led to a robust PABPN1-dependent hyperadenylation of a large pool of nuclear RNAs apparently due to the uncoupling of hyperadenylation from decay . Finally , we observed that a CPSF-independent poly ( A ) tail initiates PPD , but hyperadenylation was not sufficient for PPD in the absence of PABPN1 . From these studies , we conclude that PPD is a major human nuclear RNA decay pathway .
We aimed to generate a high-confidence list of PPD targets by performing RNA-seq on polyadenylated RNA from cells in which PPD-associated hyperadenylation had been inactivated by two independent methods . For one treatment , we prepared RNA from cells after a three-day co-depletion of PAPα and PAPγ by siRNAs ( siPAP ) ( S1A Fig ) . For the second treatment , we created a stable cell line expressing myc-tagged LALA under control of a tetracycline-responsive promoter ( TetRP ) . Following a three-day induction of LALA , we collected RNA in preparation for high-throughput sequencing . Under these conditions , LALA was expressed at levels only slightly greater than endogenous wild-type PABPN1 ( S1A Fig ) . We examined polyadenylated RNAs from detergent-insoluble nuclear fractions of control , LALA , and siPAP-treated cells and on total polyadenylated RNA from control and siPAP-treated cells . Our fractionation procedure enriches for chromatin and nuclear speckle-associated RNAs [18–20] . Admittedly , the protocol results in the loss of some detergent-soluble nuclear material , but the fractions have little cytoplasmic contamination . We identified 1339 differentially expressed genes ( DEGs ) with increased ( upregulated ) and 1576 DEGs with decreased ( downregulated ) levels in at least one PPD inactivation condition ( Fig 1A ) ( S1 Table ) . We defined high-confidence PPD targets to be those DEGs upregulated in all three datasets ( Fig 1B and 1C ) . Interestingly , 39% ( 138/353 ) of the high-confidence transcripts mapped to unannotated loci in the reference genome , while only one of the 131 overlapping downregulated genes ( 0 . 8% ) was unannotated . We visually inspected the sequence traces of all high-confidence transcripts and categorized them as mRNAs or one of several classes of ncRNA: promoter upstream transcripts ( PROMPTs , also known as TSSa-RNAs ) [21 , 22] , antisense RNAs ( AS ) , primary miRNA ( pri-miRNA ) , ncSNHG , or lncRNA ( S2 Table ) . Most of the RNAs were ncRNAs ( 80% , Fig 1D ) . We additionally performed an independent bioinformatic analysis utilizing a dataset including nearly 14 , 000 known and novel annotated lncRNAs ( GENCODE ) ( see Materials and Methods ) . For this analysis , we observed 1178 upregulated lncRNA DEGs in at least one PPD inactivation condition and 408 of these were identified in all three data sets ( S1C–S1E Fig and S3 Table ) . Thus , a considerable number of noncoding polyadenylated nuclear RNAs accumulate upon LALA overexpression and PAP knockdown , suggesting that these transcripts are PPD substrates . Eukaryotic promoters produce bidirectional transcripts , but generally only one direction produces a stable RNA [22–25] . With respect to number and fold change , PROMPTs were the most responsive class of PPD targets ( Fig 1D–1F ) . Importantly , composite RNA profiles confirmed that our visual assignment of PROMPT was accurate ( Fig 1G and S1B Fig ) . Interestingly , we observed a small peak upstream of the transcription start site ( TSS ) when the entire genome was used for the composite ( dotted lines ) , suggesting an effect beyond our high-confidence targets ( solid lines ) . We validated the response of six PROMPTs to several PPD inactivation strategies ( Fig 1H ) . In addition to LALA expression and PAP knockdown , we knocked down PABPN1 ( siPABPN1 ) , or co-depleted the two catalytic components of the exosome , DIS3 and RRP6 ( siExo ) ( S1A Fig ) . We also inhibited poly ( A ) tail extension using cordycepin , an adenosine analog that acts as a chain terminator for poly ( A ) polymerase due to the absence of a 3´ hydroxyl group . As expected , the levels of the PROMPTs increased upon PPD inactivation , but in some cases PABPN1 knockdown did not have an effect . This is likely due to a general impairment of transcription upon PABPN1 depletion ( see below ) . We previously reported that an intronless β-globin reporter RNA is degraded by PPD , but its spliced counterpart is stable [5] . Therefore , we tested whether there was a correlation between number of exons and PPD susceptibility . We found that upregulated genes had significantly fewer exons ( median 2 ) than genes from the reference list ( median 7 ) , or downregulated DEGs ( median 8 ) ( Fig 1I ) . Noncoding RNAs tend to have fewer exons than protein-coding mRNAs , so our results could be explained by the high proportion of ncRNAs in our dataset , rather than a direct consequence of reduced number of exons . However , even mRNA targets had significantly fewer exons than genes from the reference list ( median 3 vs median of 7 , p<0 . 0001 ) . Moreover , the fold changes upon PPD inactivation inversely correlated with the number of exons ( S2 Fig ) . We conclude that PPD substrates have on average fewer exons than transcripts that are not targeted by PPD . Nonetheless , a number of decay targets are spliced , demonstrating that a single splicing event is not always sufficient to confer resistance to PPD . While mRNA targets had significantly fewer exons than the reference genes , the mRNA targets had more exons than other PPD target categories except ncSNHGs ( Fig 1J ) . Interestingly , mRNAs also had a significantly lower fold-change upon PPD inactivation ( Fig 1F ) and mRNAs were expressed at higher basal levels than all other classes except ncSNHGs ( S1F and S1G Fig ) . These data suggest within the cellular pool of the specific PPD-susceptible mRNAs , a subset is exported and thereby escapes PPD . As a result , the mRNAs are less affected by PPD inactivation than PROMPTs , which are presumably not exported . Most mammalian snoRNAs are excised from introns , but the host genes can produce either coding or noncoding RNAs [26] . We identified several ncSNHGs in our RNA-seq analysis and additional ncSNHGs were upregulated that did not meet our stringent cutoffs . In order to obtain a more complete list of ncSNHG PPD targets , we performed qRT-PCR on 24 ncSNHGs expressed in our cell line following inactivation of PPD by several independent methods . In addition , we inactivated NMD by cycloheximide treatment , which indirectly inhibits NMD by inhibiting translation , or by knocking down the NMD factor UPF1 ( S1A Fig ) . Strikingly , we observed largely non-overlapping clusters of ncSNHGs targeted by NMD or PPD ( Fig 2A ) . No upregulation was observed when we used primers that detect the intron-containing transcripts ( Fig 2B ) , so PPD targets the spliced product . We next examined the effects of inactivating both pathways simultaneously . We reasoned that ncSNHGs that evade PPD in the nucleus may be exported and degraded by NMD in the cytoplasm . However , simultaneous PAP knockdown and cycloheximide treatment did not lead to additive accumulation of PPD targets ( Fig 2C ) , suggesting that NMD does not simply degrade ncSNHGs that escape PPD . Instead , each ncSNHG is targeted by a specific pathway . Consistent with our observation that the number of exons inversely correlates with PPD susceptibility , intron-poor ncSNHGs were more likely to be targeted by PPD ( Fig 2D ) . Because NMD and PPD function in the cytoplasm and nucleus , respectively , and splicing promotes mRNA export [27] , we reasoned that differences in ncSNHG localization may contribute to PPD-sensitivity . To test this hypothesis , we calculated a nuclear enrichment score ( NES ) by dividing the fragments per kilobase of exon per million reads mapped ( FPKM ) in the nuclear dataset by the FPKM value in the total dataset for each expressed gene . Plotting the NESs confirmed that the nuclear lncRNA MALAT1 had a high NES ( blue ) , while ACTB and RPL30 mRNAs received lower scores ( red ) ( Fig 2E ) . Next , we compared the NES to the fold changes observed upon PPD inactivation and found that PPD targets were typically more nuclear , while non-PPD targets were more cytoplasmic ( Fig 2F ) . Thus , the differences in nuclear retention and number of exons influence susceptibility to PPD . The simplest interpretation of these results is that fewer splicing events lead to less efficient nuclear export , which in turn increases PPD-susceptibility . MAT2A is a high-confidence PPD target and inspection of its sequence traces revealed retention of the 3´-most intron ( Fig 3A ) . Recent studies have established that intron retention is significantly more common in mammals than previously appreciated [28–31] . Retained intron-containing RNAs ( RI-RNAs ) can be degraded by NMD , but most are degraded in the nucleus by an unknown pathway [28 , 29 , 31] . We tested whether PPD affects RI-RNA decay more generally by examining MAT2A and two other RI-RNAs , OGT and ARGLU1 . Each gene produced highly expressed nuclear RI-RNAs and fully spliced cytoplasmic mRNAs ( Fig 3A and 3B ) . The presence of the retained intron is verified below ( Fig 4A ) . Neither ARGLU1 nor OGT was identified as a high-confidence target , but ARGLU1 was upregulated in the siPAP-total and siPAP-nuclear datasets . Similarly , cordycepin treatment increased MAT2A-RI levels ~2-fold , but this effect did not reach statistical significance ( p = 0 . 10 ) and cordycepin did not affect OGT-RI or ARGLU1-RI levels ( Fig 3D ) . While these data suggest little PPD sensitivity , none of the RI-RNAs responded to UPF1 depletion and only OGT-RI increased in response to cycloheximide , consistent with previous reports that NMD is not the general mode of decay for these RNAs [28 , 29 , 31] . To further probe a potential role of PPD in RI-RNA decay , we tested whether timing of the knockdown experiments influenced our results . When we increased siPAP treatment from three to four days , we observed statistically significant upregulation of MAT2A-RI ( 4 . 2-fold ) , OGT-RI ( 2 . 5-fold ) , and ARGLU1-RI ( 2 . 5-fold ) supporting the conclusion that PPD targets RI-containing RNAs ( Fig 3C and 3D ) . PABPN1 knockdown increased ARGLU1-RI levels ~1 . 8-fold , but neither MAT2A-RI nor OGT-RI increased ( Fig 3C and 3D ) . Unlike siPAP treatments , extended knockdown of PABPN1 did not increase RI-RNAs . Moreover , the cell morphology was generally worse for PABPN1 knockdowns compared to PAP knockdowns suggesting greater toxicity . Therefore , we hypothesized that decreases in transcription prevent accumulation of RI-RNAs upon PABPN1 depletion . To test this idea , we performed nuclear run-on ( NRO ) assays using the modified nucleotide , 4-thiouridine triphosphate ( 4SUTP ) , to detect nascent transcripts . We observed a general decrease in Pol II density on several genes after PABPN1 knockdown ( Fig 3E ) . We conclude that steady-state levels of some PPD targets do not increase upon PABPN1 knockdown due to concomitant decreases in RNA synthesis rates . Importantly , we detected no change in transcription upon PAP knockdown ( S3C Fig ) , consistent with our observation that RI-RNAs accumulate after PAP knockdown . We further corroborated the NRO results by examining nascent transcripts from live cells using a metabolic labeling protocol ( S3E Fig ) . These results support a role for PPD in degradation of nuclear RI-RNAs but suggest that the relative rates of transcription and decay of RI-RNAs may differ from the more robustly upregulated ncRNAs such as PROMPTs . We also examined the mRNA isoform of MAT2A , OGT , or ARGLU1 , and observed no general trends ( S3B Fig ) . We suggest this is due to distinct half-lives , translation efficiencies , and/or the precursor-product relationship between a specific RI transcript and its cognate mRNA . Initially , we attempted to examine MAT2A-RI stability by treating cells with the general transcription inhibitor actinomycin D ( ActD ) . As expected , the mRNA degraded over time ( Fig 4A ) . Surprisingly , the MAT2A-RI isoform was robustly hyperadenylated upon ActD treatment and the transcript persisted . We verified that this transcript corresponded to the MAT2A-RI by stripping and re-probing with a retained-intron specific probe ( lanes 7–12 ) . In addition , ARGLU1-RI and the OGT-RI transcripts were stable and hyperadenylated after ActD treatment ( Fig 4A ) . Because these transcripts are longer than MAT2A-RI , the hyperadenylation was not as obvious as for MAT2A . Therefore , we cleaved the transcripts ~500 nt from their 3´ ends using RNase H and a specific targeting DNA oligonucleotide and examined the 3´ fragment prior to and after ActD treatment ( Fig 4B ) . Hyperadenylated and shorter poly ( A ) tails were readily detected , reflecting the RI and mRNA isoforms , respectively . After ActD treatment , the hyperadenylated tails ranged from ~300–800 nt , while mRNAs were ~50–200 nt ( S4 Table ) . NEAT1 , a known ncRNA PPD target [4] , was also hyperadenylated after ActD treatment ( Fig 4A , lanes 13–18 ) . In contrast , neither β-actin nor GAPDH mRNAs displayed poly ( A ) tail extension upon ActD treatment ( lanes 13–18 ) . Moreover , the nuclear ncRNA MALAT1 , which does not have a poly ( A ) tail [32] , was not extended upon ActD treatment . MAT2A and ARGLU1 RNAs of intermediate lengths were hyperadenylated after ActD treatment ( Fig 4A , asterisks ) . We observed only two bands corresponding to fully spliced and RI-RNAs after RNase H/oligo ( dT ) treatment , so we conclude that these RNAs are spliced , but still subject to hyperadenylation and nuclear retention . ( S3D Fig ) . We discuss possible mechanisms of production of these RNAs in the Discussion section . PABPN1 knockdown prevents the hyperadenylation of RI-RNAs after ActD treatment ( Fig 4C , compare lanes 2 with 4 ) . PABPN1 depletion also decreased the length of MAT2A-RI in the untreated samples ( lanes 1 and 3 ) , but the MAT2A mRNA lengths were largely unaffected . Similar results were observed with PAP knockdown ( Fig 3C ) . Thus , PABPN1 and PAP hyperadenylate MAT2A-RI even in control cells and similar results were observed with ARGLU1-RI and OGT-RI isoforms ( Fig 4C ) . If PABPN1 knockdown released RI-RNAs from the nucleus , the shorter poly ( A ) tails could be due to cytoplasmic deadenylation . However , the RI-RNAs remained predominantly nuclear upon PABPN1 depletion ( S3A Fig ) . We conclude that RI-containing transcripts have longer poly ( A ) tails due to PABPN1 and PAP activity , and that this effect is exacerbated following treatment with ActD . MAT2A-RI is targeted by PPD , but upon ActD treatment the poly ( A ) tail is extended and the RNA is relatively stable . One interpretation of this finding is that ActD treatment decouples hyperadenylation from decay . To test this with a different PPD target , we compared the half-lives of SNHG19 after ActD treatment with a 4SU metabolic pulse-chase assay that does not require general transcription inhibition ( Fig 4D ) . The apparent half-life of SNHG19 in ActD was >3hr , while the pulse-chase method yielded a <30 min half-life ( Fig 4E ) . These observations show that some PPD targets are stabilized by general transcription inhibition and highlight the potential caveats of using general transcription inhibitors to monitor nuclear RNA half-lives . To explore the generality of the ActD-induced hyperadenylation , we collected RNA from cells treated with ActD over a 6-hr time course and digested them with RNase T1 , a G-specific endonuclease , to degrade transcripts but leave poly ( A ) tails intact . We then detected bulk poly ( A ) tails by northern blot with an oligo ( dT ) 40 probe ( Fig 5A ) . After ActD treatment , one subset of poly ( A ) tails lengthened , while another population shortened over time . We observed similar effects with 5 , 6-dichloro-1-β-D-ribofuranosylbenzimidazole ( DRB ) , flavopiridol , and triptolide , which inhibit transcription by mechanisms distinct from ActD ( S4A Fig ) [33] . Moreover , this hyperadenylation was observed in HeLa cells and primary mouse macrophages , so the effect is neither cell-type nor species-specific ( S4B Fig ) . Admittedly , the fraction of RNAs hyperadenylated is lower than its appearance on the northern blots ( Fig 5A ) because more oligo ( dT ) 40 probes will hybridize to the longer tails to increase the signal , but the hyperadenylated transcript pool nonetheless comprises a large fraction of the total poly ( A ) RNA . The two bulk poly ( A ) pools closely mimicked our observations with RI-RNA and mRNA isoforms . For example , the shorter population was primarily cytoplasmic whereas the hyperadenylated RNAs were nuclear ( Fig 5B ) . Moreover , the poly ( A ) tails were longer in the nuclear pool even in the absence of ActD and hyperadenylation was diminished in PABPN1-depleted cells ( Fig 5C ) . Next , we used a metabolic pulse-chase assay to examine bulk poly ( A ) tail dynamics ( Fig 5D ) . As expected , the cytoplasmic poly ( A ) tails shortened over time and ActD did not appreciably change this pattern ( Fig 5E ) . In the absence of ActD , the nuclear poly ( A ) tails grew longer but disappeared over time . In contrast , in the presence of ActD , the nuclear poly ( A ) tails persisted and were continually extended , thereby mirroring the hyperadenylation and lack of nuclear decay observed with specific PPD substrates ( Fig 4 ) . We conclude that a large fraction of nuclear polyadenylated RNA is subject to hyperadenylation and stabilization upon general transcription inhibition . PABPN1 and PAPα/γ are components of the 3´-end formation machinery , but whether other components , like CPSF , are involved in PPD is unknown . Even though hyperadenylation occurs after the initial polyadenylation event , CPSF may remain bound to the PAS and influence hyperadenylation or decay . To test this , we took advantage of the unusual processing of the MALAT1 lncRNA . The MALAT1 3´ end is generated by RNase P , which cleaves directly upstream of a tRNA-like element in the RNA [32] . We cloned the tRNA-like element into a TetRP-driven ENE-lacking PAN RNA reporter immediately downstream of a 35-nt A stretch ( Fig 6A ) ( PANΔENE-A35 ) . The processing at the MALAT1 cleavage site is efficient , with ~85% of the RNAs being cleaved by RNase P after a 2-hr transcription pulse ( S5A Fig ) . In cells , the A35 tail was extended to ~100–500 nt ( Fig 6B ) . Importantly , the cleaved transcript lacks an AAUAAA site , so this extension was independent of CPSF . To examine PANΔENE-A35 stability , we used a TetRP-based transcription pulse-chase strategy . After a 2-hr transcription pulse , we monitored stability of PANΔENE-A35 and PANΔENE with its natural PAS ( PANΔENE-AAUAAA ) and observed indistinguishable decay kinetics ( Fig 6C and 6D ) . Moreover , knockdown of PABPN1 ( Fig 6E and 6F ) or LALA expression ( S5B Fig ) stabilized PANΔENE-A35 . Thus , PPD does not strictly require CPSF or a PAS . PABPN1 , but not CPSF , stimulates polyadenylation after the initial processive polyadenylation step by increasing PAP association with RNA [13] . We previously proposed that this in vitro activity reflects the hyperadenylation required for PPD , which is further supported by the demonstration that PPD can occur in a CPSF-independent fashion ( Fig 6A–6F ) . In principle , stimulation of hyperadenylation could be the sole requirement for PABPN1 in PPD . To test this hypothesis , we bypassed the requirement for PABPN1 in hyperadenylation by tethering PAP directly to PANΔENE RNA . We inserted six bacteriophage MS2 coat protein binding sites into PANΔENE upstream of the poly ( A ) tail , which allows us to tether an MS2-PAP fusion protein to PAN RNA in cells ( PANΔENE-6MS2 ) ( Fig 6G ) . When MS2-binding protein was expressed , PANΔENE-6MS2 was rapidly degraded in control cells ( Fig 6H , lanes 5–8 ) , but stabilized upon PABPN1 knockdown ( Fig 6H , lanes 13–16 ) . When we co-expressed PANΔENE-6MS2 with MS2-PAP , PANΔENE-6MS2 was rapidly degraded in control cells as expected ( Fig 6H , lanes 1–4 ) . Importantly , MS2-PAP was unable to rescue decay after PABPN1 depletion , despite the fact that PANΔENE-6MS2 was hyperadenylated ( Fig 6H , lanes 9–12 ) . Therefore , hyperadenylation is not sufficient to stimulate PPD in the absence of PABPN1 , suggesting that PABPN1 serves multiple functions in PPD by promoting hyperadenylation and an additional step in RNA decay .
The mechanisms and regulation of nuclear RNA decay remain poorly defined , particularly in mammalian cells . Here we show that several classes of nuclear noncoding RNAs are subject to degradation by PPD including upstream antisense RNAs , ncSNHGs , pri-miRNAs , lncRNAs , and antisense transcripts . Our observations are consistent with global analyses reported by Bachand and colleagues demonstrating that PABPN1 knockdown leads to the stabilization of nuclear lncRNAs [4] . In addition , our RNA-seq and knockdown analyses revealed that specific canonical mRNAs and RI-containing RNAs are PPD targets . By using PAP knockdown and PAP-stimulation deficient PABPN1 mutant LALA as the basis of our RNA-seq experiments , these data confirm that PAP activity is necessary for the degradation of a large collection of nuclear RNAs . Given the parameters used in the RNA-seq analysis , it is likely that our high-stringency dataset is an underestimate of the number of RNAs subject to PPD . For example , a subset of ncSNHGs and the RI-RNAs were confirmed to be PPD substrates by qRT-PCR ( Fig 2A ) and northern blot ( Fig 3C and 3D ) even though these RNAs were not identified in our RNA-seq study . Based on these global and mechanistic studies we conclude that PPD is a major RNA decay pathway for nuclear polyadenylated transcripts . The PROMPTs were the most PPD sensitive transcripts based on their fold changes upon PPD inactivation ( Fig 1F ) and their overrepresentation among DEGs ( Fig 1D ) . Pervasive transcription from bidirectional promoter firing is a common feature in eukaryotes [1 , 22 , 23 , 25 , 34 , 35] . In S . cerevisiae , the resulting divergent transcripts are terminated by the Nrd1-Nab3-Sen1 ( NNS ) pathway due to an over-representation of binding sites for the Nrd1p and Nab3p proteins upstream of yeast promoters [36 , 37] . The multisubunit Trf4-Air2-Mtr4 polyadenylation ( TRAMP ) complex then targets the NNS-terminated fragments to the nuclear exosome [38–40] . In contrast , promoter directionality in mammalian cells is achieved by an enrichment in canonical PASs in the upstream antisense direction and depletion of U1 snRNP binding sites [41 , 42] . At least some PROMPTs are terminated by the combined actions of the canonical cleavage and polyadenylation machinery , the cap-binding complex and its associated protein ARS2 [41–44] . After termination , the trimeric NEXT complex targets PROMPTS for decay by the exosome [24 , 43 , 45 , 46] . In addition , bidirectional transcripts can be terminated and degraded by co-transcriptional decapping and 5´→3´ decay by Xrn2 [47] . Three studies , including this one , report that specific PROMPTs are degraded in a PABPN1-dependent fashion [4 , 48] . Visual inspection of the sequence traces of previously published NEXT-sensitive PROMPTS is ambiguous regarding their susceptibility to PPD ( S6A Fig ) [4] , suggesting that specific PROMPTs are targeted by distinct nuclear decay pathways . Further experimentation is required to determine whether the PPD , Xrn2 and NEXT pathways target independent subsets of upstream antisense transcripts , or are largely redundant pathways for bidirectional transcript degradation . U1 snRNP is a core component of the spliceosome that recognizes 5´ splice sites , but it also suppresses the use of premature PASs [49 , 50] . This latter function contributes to promoter directionality in that U1 snRNP binding sites are depleted in upstream antisense regions and overrepresented in coding regions [41 , 42] . As a result , antisense transcription normally produces shorter , unspliced transcripts , whereas coding genes produce longer spliced pre-mRNAs . Interestingly , five of our high-confidence PPD substrates classified as mRNAs had increased sequence coverage at the 5´ end of the genes ( APOLD1 , MTHFD2L , AGBL3 , TEX22 , and FAM120C ) ( S6B Fig ) . We speculate that these transcripts result from a failure of U1 snRNP to protect from premature PAS usage . The resulting RNAs resemble promoter antisense RNAs and are therefore subject to degradation by PPD . This speculation is supported by a recent global analysis demonstrating that PABPN1 depletion increased the levels of similar sense proximal RNAs [48] . We previously demonstrated that an intronless β-globin reporter is rapidly degraded by PPD , but insertion of a single intron into that reporter is sufficient to protect the resulting mRNA from PPD [5] . Consistent with this idea , 174/353 ( 49% ) of the high-confidence RNAs identified are single-exon RNAs ( S2 Fig ) . The simplest explanation for this observation is that splicing promotes the formation of an export-competent mRNP leading to export and escape from PPD [27] . However , a single splicing event is not always sufficient to promote escape from PPD . By definition , all PPD-targeted ncSNHGs are spliced at least once ( Fig 2 ) and only 5/74 PPD-sensitive mRNAs are single exon genes ( S2 Table ) . Because ncSNHGs targeted by PPD had higher nuclear enrichment ( Fig 2 ) , we conclude that PPD susceptibility stems from nuclear retention of the spliced transcript . This could be due to nuclear retention signals in the exons or due to variations in recruitment of splicing-dependent export factors . We also found that RI-RNAs are subject to PPD ( Figs 3 and 4 ) . Recent studies point out the importance of intron retention in mammalian cells [28–31] . The efficiency of splicing of these retained ( “detained” in [31] ) introns can be modulated by developmental or environmental cues supporting an essential role for these RNAs in posttranscriptional gene regulation . These previous studies showed that a subset of RI-RNAs is degraded by NMD while others are retained in the nucleus and degraded by a previously unknown nuclear RNA decay pathway . Our data now show that that nuclear retained RI-RNAs are subject to PPD . Thus , there is a parallel between RI-RNAs and ncSNHGs in that both produce spliced RNAs that are either exported and subject to NMD or retained in the nucleus and subject to PPD . Importantly , the RI-RNAs are not strongly upregulated by PPD inactivation . We had to increase the lengths of time for PAP knockdown to observe increases in ARGLU1 and OGT and cordycepin treatment had no effect on their abundance ( Fig 3D ) . This may be due to the biology of the RI-RNAs . For example , if they serve as precursors to pre-mRNAs as proposed [31 , 51] , the half-lives of these RNAs may be longer than the nonfunctional ncSNHGs or PROMPTs . Thus cells may regulate PPD to control the accumulation of RI-RNAs . Given the widespread use of intron retention in mammals , PPD regulation may have important consequences for gene expression . Interestingly , PABPN1 was recently shown to autoregulate its mRNA levels by intron retention [52] . Testing the half-lives of the nuclear RNAs identified herein is complicated by the unusual behavior of nuclear RNAs upon general transcription inhibition ( Figs 4 and 5 ) . We do not understand how transcription inhibition leads to the accumulation of hyperadenylated nuclear RNAs , but the simplest explanation for this striking phenomenology is that PABPN1-dependent hyperadenylation occurs , but is uncoupled from the decay step of PPD . We stress that this is not the result of a specific transcription inhibitor or concentration as four different transcription inhibitors , which utilize at least three distinct mechanisms of transcription inhibition yielded a similar result ( S4 Fig ) . Interestingly , we observed that a portion of completely spliced MAT2A and ARGLU1 RNAs was hyperadenylated after ActD treatment ( Fig 4A and S3D Fig ) . Because there is little fully spliced RNA in the nuclear fraction prior to ActD treatment ( S3D Fig ) , it seems likely that the retained intron is posttranscriptionally spliced . However , this splicing is not sufficient to release the RNA for export , at least in the presence of ActD . Perhaps transcription inhibitors indirectly produce a general block in mRNA export . Alternatively , the RI-RNAs may be fated for the discard pathway , so they are subject to nuclear retention and PPD even after splicing . Another explanation is that the RI-RNAs are normally degraded , but ActD-induced stabilization ( Figs 4 and 5 ) allows sufficient time for the RNAs to be fully spliced . Given the prevalence of intron retention in mammals , the interrelationships between PPD , splicing , and transcription warrant deeper investigation . In yeast , the TRAMP complex component Trf4 , a noncanonical poly ( A ) polymerase , marks nuclear RNAs for decay by the exosome . While Trf4 is essential for decay , its polyadenylation activity is not necessary [53–55] . In contrast , our studies are consistent with the conclusion that hyperadenylation of PPD targets is linked to their decay . Transcripts that are upregulated following PABPN1-depletion are also increased following depletion of PAP or expression of a polyadenylation defective PABPN1 allele ( Figs 1 and 2 ) . Three lines of evidence suggest that distributive rather than processive polyadenylation is the primary driver of decay . First , CPSF is necessary for processive polyadenylation in vitro so the CPSF-independent PANΔENE-A35 is unlikely to undergo processive polyadenylation . Nevertheless , PANΔENE-A35 was degraded by PPD ( Fig 6 ) , suggesting that processive polyadenylation is dispensable for decay . Second , a distributive process should be more sensitive to relative concentrations of PPD factors in the cell because of the requirement for re-binding after dissociation . Indeed , our siPAP knockdowns decrease PAP levels such that hyperadenylation is affected , but there appears to be little effect on the initial polyadenylation reaction [5] . Third , upon transcription inhibition , poly ( A ) tails gradually increased in length as a group over several hours , consistent with PAP disassociating and re-associating with transcripts stochastically ( Figs 4 and 5 ) . In contrast , processive polyadenylation that forms the initial poly ( A ) tail occurs rapidly in vitro and in cells with ~200–250 nucleotides being added in less than one minute [56 , 57] . Interestingly , even though PABPN1 stimulates CPSF-independent distributive hyperadenylation , hyperadenylation was not sufficient to rescue PPD sensitivity in the absence of PABPN1 ( Fig 6H ) . Thus , PABPN1 likely plays multiple roles in PPD . In fact , Pab2 and PABPN1 co-immunoprecipitate with the exosome [4 , 58] , suggesting PABPN1 may directly recruit the exosome . Alternatively , PABPN1 may compete with poly ( A ) binding proteins that stabilize RNAs . Thus , upon PABPN1 depletion , these proteins preferentially associate to increase RNA half-lives [59 , 60] . In summary , our data show that PPD modulates the levels of functional lncRNAs and mRNAs as well as presumably nonfunctional PROMPTs and the spliced byproducts of snoRNA and pri-miRNA processing . We conclude that PPD is an important nuclear RNA decay pathway that lies at the interface of transcription , splicing , 3´-end formation and mRNA export .
RNA-seq and sequencing was performed at the McDermott Center Next Generation Sequencing Core and Bioinformatics Core . Libraries were prepared using the TruSeq Stranded mRNA preparation kit and run on an Illumina HiSeq 2500 ( paired-end 100 bp reads ) . The reads were mapped , aligned and assembled using TopHat2 and Cufflinks2 . 2 [61 , 62] . Transcriptome assembly was guided by iGenomes ( hg19 , UCSC build ) and GENCODE ( release 19 ) annotation files . Differential gene expression was analyzed by Cuffdiff using the iGenomes annotations and EdgeR was employed to determine differential expression of the 13 , 853 known and novel lncRNAs in the GENCODE annotation [63] . Integrative genomics viewer ( IGV ) was used to visualize sequence coverage and generate figures [64] . DEGs were identified from the Cuffdiff output by removing those transcripts with an FPKM of <1 in the treatment sample and the remaining transcripts with p-value <0 . 05 and a false discovery rate ( FDR ) less than 5% were defined as DEGs ( S1 Table ) . DEGs in the EdgeR data were defined as those with log ( counts per million ) >3 . 5 and an FDR <5% ( S3 Table ) . Heat maps were generated using the GENE-E software ( http://www . broadinstitute . org/cancer/software/GENE-E/index . html ) . We categorized each of the 353 high-confidence upregulated DEGs by visual assessment of IGV traces ( S2 Table ) . Any DEG found upstream and antisense to an annotated gene was defined as a PROMPT . Antisense orientation was confirmed in IGV using strand-specific bigWig files generated by HOMER [65] . AS transcripts , on the other hand , were those with considerable overlap within an annotated gene . Pri-miRNA and ncSNHG transcripts were inferred by the presence of an overlapping miRNA/snoRNA or corresponded to annotated genes . We assigned the category lncRNA to any transcript that was from an annotated lncRNA gene or from an unannotated genomic region that did not fall into any of the other categories . All plasmids were constructed using standard molecular biology techniques . The details of the construction are given in the Expanded View . Transfections and TetRP pulse-chase assays were performed as previously described [5] . Detection of newly made bulk poly ( A ) tails was performed essentially as previously described [66 , 67] . Bulk poly ( A ) tails were detected on 1 . 8% agarose-formaldehyde gel , and detected with a dT40 probe end-labeled with T4 polynucleotide kinase . Northern blots for specific transcripts were performed using standard techniques with RNA probes . Stripping and re-probing of the membranes were performed as previously described [5] . The RNA probes were generated from PCR products with a T7 RNA polymerase promoter; primers are listed in S5 Table . For some northern blots , 35–80 mg of total RNA were selected on oligo ( dT ) cellulose to enrich for polyadenylated RNAs prior to gel electrophoresis . In addition , we degraded residual rRNA after oligo ( dT ) -cellulose selection with Terminator exonuclease ( EpiCentre ) . To collect cytoplasmic RNA , cells were resuspended in Buffer I ( 0 . 32 M sucrose , 3mM CaCl2 , 2 mM MgCl2 , 0 . 1 mM EDTA , 10 mM Tris-HCl ( pH 8 . 0 ) , 1 mM DTT , 0 . 04 U/ml RNase Inhibitor , 0 . 5% Triton X-100 ) , incubated on ice for 5 min , centrifuged at 500 x g for 3 min at 4° . RNA in the supernatant was extracted using TriReagent ( Molecular Research Center ) followed by an additional phenol-chloroform extraction . The pellet was then washed in Buffer I with 150 mM NaCl and once again centrifuged at 500 x g for 3 min at 4° . The resulting supernatant was discarded . The RNA from the remaining pellet was then extracted in TriReagent . We note that in cases in which we analyzed RNA from the wash step , we observed both long and short poly ( A ) tails; whether this is due to cross contamination of cellular compartments and/or is due to a distinct biological fraction is unclear . This fractionation procedure results in the loss of Triton X100-soluble nuclear material , but it enriches for chromatin and nuclear speckle-associated RNAs [18–20] . RNA was harvested using TriReagent according to the manufacturer’s protocol . Following extraction , RNA was treated with RQ1 DNase ( Promega ) . Random hexamers were used to prime cDNA synthesis with MuLV reverse transcriptase ( NEB ) . Real-time reactions used iTaq Universal SYBR Green Supermix ( Biorad ) . Biotinylation reactions were carried out in a 200μL mixture consisting of 40μg RNA , 20mM NaOAc ( pH 5 . 2 ) , 1mM EDTA , 0 . 1% SDS , 0 . 2mg/mL Biotin-HPDP ( Pierce ) , and 50% N , N-dimethylformamide ( DMF ) for 3 hours at 25°C . Unconjugated biotin-HPDP was removed with three chloroform extractions . After extraction of the aqueous phase , 20μL ( 10% v/v ) of 10M NH4OAc was added to each tube , and the RNA was precipitated in 70% ethanol . Streptavidin selection was carried out using magnetic Streptavidin T1 beads ( Invitrogen ) . Prior to use , the 20 μl bead slurry was washed three times in a 0 . 1X MPG solution ( 1X MPG was 1M NaCl , 10mM EDTA , and 100mM Tris 7 . 5 ) supplemented with 0 . 1% igepal . After the final wash , the beads were resuspended in a 1mL solution consisting of 0 . 1X MPG supplemented with 0 . 1% igepal , 0 . 1μg/μL poly ( A ) ( Sigma-Aldrich ) , 0 . 1μg/μL ssDNA , 0 . 1 μg/μL cRNA , and 0 . 1% SDS , and blocked for one hour . RNA was precipitated , resuspended in a volume of 63μL water , and denatured at 65°C for 5 minutes . Next , RNA was incubated together with beads for one hour while nutating at room temperature . Beads were sequentially washed in: 0 . 1X MPG , 0 . 1X MPG at 55°C , 0 . 1X MPG , 1X MPG , 1X MPG , 0 . 1X MPG , 1X MPG without NaCl , 0 . 1X MPG . With the exception of the 55°C wash , each solution included 0 . 1% igepal . Biotinylated RNAs were eluted twice for 5 minutes each in a 200μL solution of 0 . 1X MPG containing 5% β-mercaptoethanol . The first elution step was at 25°C and the second was at 65°C . The two eluted fractions were combined and extracted with PCA once and chloroform twice . After extraction , 40μL of 10M NH4OAc was added to each tube , and the RNA was precipitated in 70% ethanol . Nuclear run-ons were performed essentially as previously described [67] . The details are provided in the Expanded View . Following knockdown , cells were treated with 2μM of 4SU for one hour . Afterwards , cells were washed twice with phosphate buffered saline ( PBS ) containing calcium and magnesium ( Sigma-Aldrich ) , and grown in media lacking 4SU for an additional hour . After the one-hour washout step , we collected 0 , 30 , 60 , and 120 min time points . 40μg of RNA was used as input for a biotinylation and streptavidin selection as described above . Selected RNA was reverse transcribed prior to qRT-PCR analysis . β-actin was used as a loading control for qPCR analysis . The cells were given fresh media 4 . 5 hours prior to the 4SU treatment , which was necessary for consistent results . Cells were treated with 100 μM of 4SU for five minutes and incorporation was quickly stopped by addition of TriReagent . Sixty micrograms of total RNA was used for biotinylation and streptavidin selection as described above except one additional 1X MPG and one additional no salt wash was performed and both elution steps were done at room temperature . | Cells control gene expression by balancing the rates of RNA synthesis and decay . While the mechanisms of transcription regulation are extensively studied , the parameters that control nuclear RNA stability remain largely unknown . Previously , we and others reported that poly ( A ) tails may stimulate RNA decay in mammalian nuclei . This function is mediated by the concerted actions of the nuclear poly ( A ) binding protein PABPN1 , poly ( A ) polymerase ( PAP ) , and the nuclear exosome complex , a pathway we have named PABPN1 and PAP-mediated RNA decay ( PPD ) . Because nearly all mRNAs possess a poly ( A ) tail , it remains unclear how PPD targets specific transcripts . Here , we inactivated PPD by two distinct mechanisms and examined global gene expression . We identified a number of potential target genes , including snoRNA host genes , promoter antisense RNAs , and mRNAs . Interestingly , target transcripts tend to be incompletely spliced or possess fewer introns than non-target transcripts , suggesting that efficient splicing allows normal mRNAs to escape decay . We suggest that PPD plays an important role in gene expression by limiting the accumulation of inefficiently processed RNAs . In addition , our results highlight the complex relationship between ( pre- ) mRNA splicing and nuclear RNA decay . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [] | 2015 | Canonical Poly(A) Polymerase Activity Promotes the Decay of a Wide Variety of Mammalian Nuclear RNAs |
Copy number variations ( CNVs ) represent a large source of genetic variation in humans and have been increasingly studied for disease association . A deletion polymorphism of the gene encoding the cytosolic detoxification enzyme glutathione S-transferase theta 1 ( GSTT1 ) has been extensively studied for cancer susceptibility ( 919 studies , from HuGE navigator , http://www . hugenavigator . net/ ) . However , clear conclusions have not been reached . Since the GSTT1 gene is located within a genomic region of segmental duplications ( SD ) , there may be a confounding effect from another , yet-uncharacterized CNV at the same locus . Here we describe a previously uncharacterized 38-kilo-base ( kb ) long deletion polymorphism of GSTT2B located within a 61-kb DNA inverted repeat . GSTT2B is a duplicated copy of GSTT2 , the only paralogue of GSTT1 in humans . A newly developed PCR assay revealed that a microhomology-mediated breakpoint appears to be shared among individuals at high frequency . The GSTT2B deletion polymorphism was in strong linkage disequilibrium ( LD ) ( D′ = 0 . 841 ) with the neighboring GSTT1 deletion polymorphism in the Caucasian population . Alleles harboring a single deletion were significantly overrepresented ( p = 2 . 22×10−16 ) , suggesting a selection against alleles with both deletions . The deletion alleles are almost certainly the derived ones , because the GSTT2B-GSTT2-GSTT1 genes were strictly retained in chimpanzees . Extremely low GSTT2 mRNA expression was associated with the GSTT2B deletion , suggesting an influence of the deletion on the flanking region and loss of GSTT2 function . Genome-wide LD analysis between deletion polymorphisms further points to the uniqueness of two deletions , because strong LD between deletion polymorphisms might be very rare in humans . These results show a complex genomic organization and unexpected biological functions of CNVs within segmental duplications and emphasize the importance of detailed structural characterization for disease association studies .
Copy number variation ( CNV ) is a significant source of genetic variation in the genome of humans [1]–[11] . A large number of CNVs has been identified , and span more than 10% of the human genome in total [12] , although the estimate is dependent on the frequency of the event under consideration . The biomedical relevance of CNVs is expected to be significant , because many CNVs cover large genomic regions and include exons and regulatory elements that are important for proper cellular function . However , these CNVs are primarily identified by indirect , array-based methods with limited resolution; defining fine scale structure , especially for large CNVs , is just beginning at the sequence level [9] , [13] , [14] . Without such information , it is difficult to determine each CNV's history , population structure , and influence on the function of one or more genes within the CNV and surrounding genomic regions . CNVs are significantly enriched in the regions of segmental duplications ( SD ) [6]–[8] , [10] , [12] . SDs are highly identical DNA segments that map to two or more loci within the genome [15] , [16] . Since regions of SDs have strong positive correlations with genes [15] , [17] , CNVs that overlap with SDs are particularly gene-rich . Therefore , defining the extent and breakpoint in each CNV in regions of SD is particularly important in order to identify CNVs that may have clinical relevance . In fact , CNVs are highly enriched in gene classes such as defense and immune response [1] , [18] , suggesting a link between CNVs in SDs and human health . However , determining the detailed structures of CNVs in SDs is not an easy task . First , given the fact that DNA sequences in SDs vary substantially among individuals , any technology based on the reference genome sequence may not be sufficient to accurately map all CNVs . Second , single nucleotide polymorphisms ( SNPs ) , the most widely used markers to tag genomic locations , are not always reliable within SDs [19] , [20] . Although SNP-based methods have identified a large number of deletion polymorphisms successfully [1] , [5] , this approach may not be as efficient in SDs as within unique segments of the genome . Therefore , more direct approaches , such as clone-based sequencing for mapping breakpoints , and subsequent molecular assays for genotyping , are necessary to accurately interrogate CNVs in regions of SDs [21] . The importance of CNVs in human diseases has become increasingly apparent [22] , [23] . It has long been known that DNA rearrangements of large genomic regions play a major role in the pathogenesis of rare genetic diseases ( genomic disorders ) [24]–[26] , and more recently , more common complex diseases such as non-syndromic mental retardation , autism and schizophrenia [27]–[30] . Common deletion polymorphisms of a class of genes in cellular detoxification , glutathion S-transferases ( GSTs ) , have also been known for more than a decade [31] , [32] . GST is a supergene family . Each sub-family member is located in a distinct genomic region and consists of as many as five paralogues [33] . GST gene products catalyze the conjugation of reduced glutathione to electrophilic centers for a wide variety of substrates [34] . The increased solubility of the conjugated products renders them more readily eliminated by the cell . Substrates include both xenobiotics and endogenous compounds that are harmful to cellular macromolecules . Based on the hypothesis that lack of GST may cause reduced levels of cellular detoxification , and thus predispose individuals to common diseases such as cancer , previously defined null alleles ( deletion polymorphisms ) have been subject to extensive disease-association studies ( 1230 published studies , information obtained from HuGE Navigator ) . However , to date , the reports contain conflicting results [35]–[38] . One possible explanation for the conflict could be that due to extensive segmental duplications in the genomic loci of GST family members , there are other , yet-uncharacterized null alleles that may impact the results . In this study , using DNA samples from blood , lymphoblastoid cell lines , HapMap populations , and chimpanzees; and RNA from primary fibroblasts and cancer cell lines , we conducted a systematic genetic , gene expression and evolutionary analysis for a previously uncharacterized large deletion polymorphism located at chromosome 22q13 , a genomic region with a 61 kilo-base ( kb ) inverted repeat . Each repeat harbors a theta class of GST gene , GSTT2B on the centromeric side of the repeat and GSTT2 on the telomeric side ( Figure 1A ) . A 37-kb deletion encompassed s the entire centromeric side and the GSTT2B gene . We show here that the deletion allele is very common in all three HapMap populations . In particular , a high frequency deletion allele ( 66% ) in the CEU population is in linkage disequilibrium ( LD ) with the neighboring GSTT1 deletion polymorphism . Such a strong LD between deletion polymorphisms is indeed very rare within the currently known deletion polymorphisms . The deletion has a strong influence on the remaining GSTT2 , as we found that GSTT2 expression is severely reduced in cells with homozygous deletion of GSTT2B . SNP analysis within the deletion region , however , failed to yield null genotypes , possibly because almost all these SNPs are located within a recently duplicated region .
To identify structural variation in the regions of large DNA inverted repeats ( DNA-IR ) , we first obtained information of DNA-IRs represented in the human genome sequence ( Build 35 ) from the Inverted Repeat Database ( IRDB ) [39] . Because of secondary structures , perfect DNA palindromes , with small non-palindromic spacers between arms ( repeats ) , are predisposed to DNA rearrangements in both simple organisms and mammals [40] , [41] . Therefore , we hypothesized that large DNA-IRs with high-sequence identity between repeats and small non-palindromic spacers may often be subject to chromosome breakage and DNA rearrangement , and , as a result , likely to be enriched for structural variations . Among large DNA-IRs in the human genome , one on the chromosome 22q11 . 23 has a large repeat unit size ( 29 . 6-kb ) with 97 . 9% sequence identity between repeats , and a 2 . 1-kb spacer ( Figure 1A ) . This DNA-IR has previously been shown to be located in the region of discordance by fosmid end-mapping and copy number variation analyses [6] , [9] . Other features are also notable in this region , such as a high frequency deletion polymorphism ( GSTT1 , Figure 1A open rectangle ) , and a low density of the HapMap SNPs . The gene duplicated in the DNA-IR is GSTT2 , a theta class glutathione transferase . We use the gene name GSTT2B for the GSTT2 located on the centromeric ( left ) repeat according to the annotation in the UCSC genome browser . Molecular characterization of DNA-IRs is a challenge , because DNA-IRs with small spacers are known to be resistant to PCR amplification and cloning in E . coli . Southern analysis and restriction fragment length polymorphism has been successfully used to determine DNA structure within DNA-IRs [42] . To identify a structural variation associated with the DNA-IR , we designed a probe that was hybridized to the DNA near the non-palindromic spacer . DNA rearrangements are known to occur most frequently at the spacer and surrounding regions [43] . We also took advantage of the segmentally duplicated sequences in this locus . We designed a probe with high sequence homology to the three regions ( Figure 1B ) . By using restriction enzyme EcoRV , we could determine genotypes for both GSTT1 and GSTT2 simultaneously . EcoRV-digested genomic DNA samples of lymphoblastoid cell lines established from 38 Caucasian individuals were used to determine the lengths of three restriction fragments , including a 4 . 6-kb fragment on the telomeric ( right ) repeat of the DNA-IR , a 6 . 3-kb fragment on the centromeric ( left ) repeat , and a 16 kb fragment near the GSTT1 gene . As is shown in Figure 1B , the 6 . 3 kb fragment was very frequently missing in these samples . Nineteen samples did not have the 6 . 3-kb fragment , suggesting a homozygous deletion of the right repeat of DNA-IR . The deletion was further confirmed by using genomic DNA digested with both SfiI and NdeI ( Figure S1 ) . In addition to the potential homozygous deletion , there were samples that showed reduced intensity of the 6 . 3 kb fragment relative to the 4 . 6 kb one . These individuals could be heterozygous for the deletion . Furthermore , the 16-kb fragments were not seen in 9 individuals , suggesting a homozygous deletion of the GSTT1 gene . Finally , a unique 10 kb fragment is seen in one individual ( Figure 1B , star ) . Southern analysis above clearly illustrated a frequent deletion and complex pattern of structural variation within and near the 61-kb DNA-IR . To determine the extent and breakpoint of deletion , genome assembly comparison was performed between the NCBI Build 36 and Celera assembly ( Figure 1C ) . To identify differences at sequence-level resolution , we directly compared DNA sequences by PipMaker [44] . The DNA sequences used for this comparison cover the genomic region between MIF and GSTT1 . Self-comparison of the NCBI assembly showed a large DNA-IR that was illustrated by a large cross-line ( left ) to the main diagonal . In contrast , there was sequence discordance at the region of the DNA-IR between two assemblies ( right ) . Thirty-seven kb of genomic sequences , including an entire left repeat of the DNA-IR was missing in the Celera assembly . In fact , the DNA-IR was not seen in the dot plot created by the self-comparison of Celera assembly ( data not shown ) . In order to determine whether the frequent deletion observed by Southern analysis was represented in the Celera assembly , a PCR primer set was designed to amplify a putative breakpoint ( Figure 2A ) . This primer set amplified the 505-bp fragment from the GSTT2B deletion allele ( del ) , but could not amplify a product of 39-kb ( deleted region plus franking sequence ) from the non-deleted allele . A PCR product of expected size was seen from the individuals that show a missing or reduced intensity of a 6 . 3-kb fragment . DNA sequencing of the PCR products form 5 individuals showed that an identical breakpoint was shared among individuals . The breakpoint resided within a unique ( non-repetitive ) sequence and was mediated by 2-bp microhomology ( Figure 2B ) . From these results , we predicted that a GSTT2B-deleted allele exists at high frequency in our Caucasian samples . This allele may also be a common one in human populations , because ( 1 ) this allele is represented in the Celera assembly and ( 2 ) the breakpoint was identified by recent paired end-pair mappings with a small number of samples [13] , [14] . A 37-kb GSTT2B deletion polymorphism was located very close to another 54-kb deletion polymorphism of GSTT1 . Thus , two large , high-frequency deletion polymorphisms exist within a genomic region of 124 kb . CNVs are very common in the human genome . However , neighboring , large , high frequency deletions could be relatively rare occurrences . In order to identify whether the deletion genotype is found at a high frequency in a large sample population , we developed a PCR-based assay ( Figure 2A ) . Three primer sets were designed to simultaneously PCR-amplify both the non-deleted ( 847-bp ) and deleted allele ( 505 bp ) of GSTT2B . Similarly , previously developed PCR assay was used to detect the GSTT1 deletion [45] . These PCR-based assays were first applied to the genomic DNA from blood samples of the same Caucasian population that we used for screening by Southern analysis . To determine the robustness of our PCR-based assay to detect the GSTT2B deletion , we genotyped these samples using both Southern analysis and our PCR-based assay in a blinded manner . The results obtained by both methods were then unblended and revealed almost complete concordance ( 37/39 individuals ) . The two cases ( 2 individuals , 5% ) of discordance could be due to either the less accurate calling based on the relative intensity between the 4 . 3- and 6 . 3-kb fragments by Southern analysis , or the existence of CNV with distinct breakpoints ( Figure 1B , star ) . The frequency of the GSTT2B deletion was very high in the population analyzed; deletion allele frequency ( 0 . 54 ) was higher than that of non-deletion allele ( 0 . 46 ) ( Table 1 ) . The allele frequency of the GSTT1 deletion was 0 . 36 , which was comparable to the frequency in the CEU population ( 0 . 39 ) of the HapMap samples [5] . From the Southern analysis , we noticed a potential linkage between the two deletion polymorphisms . Individuals who did not have the 6 . 3-kb fragment tended to have the 16-kb fragment , and individuals who did not have the 16 kb fragment tended to have the 6 . 3-kb fragment . This suggests a non-random assortment ( Linkage Disequilibrium , LD ) between the two deletion polymorphisms . In order to assess LD between the deletions , we reconstructed deletion-based haplotypes using PHASE [46] ( Table 1 ) . Each deletion genotype was determined based on the results from the PCR-based assay . Haplotype frequencies at the locus were found to be significantly deviated from the expected values: single-gene deletions were overrepresented whereas alleles with both gene deletions were exceedingly rare ( p = 5 . 17×10−7 ) . The frequency of the GSTT2 deletion/GSTT1 non-deletion haplotype was 0 . 49 ( expected 0 . 34 , if random ) while the frequency of the GSTT2 non-deletion/GSTT1 deletion was 0 . 29 ( 0 . 165 , if random ) . The frequency of the haplotype with both deletions was very low , 0 . 048 ( 0 . 19 , if random ) . Thus , high frequency , neighboring deletion polymorphisms were non-randomly associated in Caucasian populations ( D′ = 0 . 7719 ) . The GSTT2B deletion was not expected to have an effect on GSTT2 expression , because the GSTT2 gene and its promoter regions were intact in the GSTT2B-deleted allele . Gene expression levels can be proportionate to the gene dosage in the case of exonic deletions [5] , in which case , we should expect a half level of GSTT2 expression . Alternatively , a large genomic deletion may influence the level of GSTT2 expression . To determine the potential effect of GSTT2B deletion on GSTT2 expression , we measured the GSTT2 mRNA expression level for each genotype . GSTT2 was not expressed at an appreciable level in the lymphoblastoid cell lines and was undetectable by Northern analysis . Therefore , we first examined 7 cancer cell lines that included three cell lines homozygous for the non-deletion allele ( HCT116 , 2008-C13 , and 2008 ) , two heterozygous ( Lovo and HCT15 ) and two cell lines homozygous for the deletion allele ( Ovaca3 and HT29 ) ( Figure 3A ) . GSTT2 expression was readily detectable in cell lines with the GSTT2B non-deletion allele . In contrast , in cell lines with homozygous deletions of GSTT2B , GSTT2 expression was undetectable ( Figure 3B ) . Cancer cell lines are very often aneuploid , which may contribute to the observed pattern of gene expression . We further determined GSTT2 gene expression using 5 primary fibroblasts . Consistent with the results from cancer cell lines , GSTT2 expression was strong in a fibroblast homozygous for the non-deletion allele , was weaker but detectable when heterozygous , and was undetectable in cell lines homozygous for the GSTT2B deletion . Finally , quantitative RT-PCR analysis ( Figure 3C ) showed relative gene expression levels that are very similar to the pattern observed for null and non-null genotype; cells homozygous for the GSTT2B deletion showed more than 80% reduction of GSTT2 expression in cell lines homozygous for the non-deletion alleles . Therefore , a large deletion including GSTT2B influences the expression of a flanking gene and correlates with the very low level of GSTT2 mRNA expression . We predicted two possible ancestral allelic states for the GSTT2B-GSTT2 region: 1 ) a single GSTT2 gene that is duplicated during the evolution of humans , or 2 ) an inverted duplication that was in part deleted in the human lineage . In principle , the ancestral allele can be inferred by analysis of the chimpanzee genome sequence assembly ( panTro2 ) . However , we were unable to determine the ancestral state due to the over-abundance of gaps surrounding the chimpanzee GSTT2 assembly . Instead , we applied molecular analyses that determined genotypes on human samples ( Figure 4 ) . Three restriction fragments representing GSTT2B , GSTT2 and GSTT1 in humans were all conserved in 12 chimpanzee samples , with an exception of a polymorphism seen in the 4 . 6-kb fragment . The results from PCR-based assays were also consistent with the non-deletion state of both GSTT1 and GSTT2B in the chimpanzee . Therefore , the ancestral state is most likely a duplicated GSTT2 , where both of the deletion alleles are derived within the human lineage . Despite its high frequency , the GSTT2B deletion polymorphism was not detectable by systematic methods using the HapMap SNP genotypes [1] , [5]; which raises the question of SNP genotypes within the DNA-IR . HapMap SNP density is lower than average within this locus: 37 SNPs within 124 kb in European ( CEU ) samples ( 1 SNP/3 . 3 kb ) ( Figure 1A ) . In order to obtain SNP genotypes within the GSTT2B deletion polymorphism , we determined the genotype of GSTT2B deletion in the HapMap samples ( Table 2 ) ( Table S1 ) . The GSTT1 deletion genotype was determined previously for the HapMap samples [5] . The frequency of the GSTT2B deletion allele was very high in CEU ( 0 . 63 ) , which is consistent with that of our Caucasian samples . The deletion polymorphism of GSTT2B spans 7 SNPs , 6 of which are located within the duplicated segment , while the GSTT1 deletion , which can be correctly identified by SNP-based methods , contains 11 SNPs ( Figure 5A ) ( Table S2 ) . For each sample , SNP genotypes were obtained from the HapMap website . We expected a null genotype ( N/N ) in case of homozygous deletion . In fact , this was the case for the GSTT1 deletion , in which two SNPs ( rs2266633 and re5760170 ) were assigned with null genotypes in more than 50% of the 15 CEU individuals with homozygous deletion . Fifteen individuals ( 100% ) were genotyped as null for rs2266633 , indicating excellent “SNP tagging” of the GSTT1 homozygous deletion . In contrast , none of the SNPs correctly genotyped the 39 individuals who are homozygous for GSTT2B deletion . One SNP ( rs9608219 ) that was located outside of the duplicated region was called as null in 5 individuals ( 11 . 6% ) , while one individual was genotyped as null for rs2330649 . None of the other SNPs were genotyped as null . Therefore , the GSTT2 deletion polymorphism status could not be genotyped correctly by the assay used for the HapMap SNP genotypes , which strongly suggests a difficulty of correctly genotyping deletions located within a recently duplicated region using SNP-based approach [19] , [20] . The GSTT2B deletion polymorphism was also very common in both the Japanese/Chinese populations ( JCP ) and the Yoruba population ( YRI ) , with an allele frequency of 0 . 50 and 0 . 47 , respectively ( Table 2 ) . Since individuals' genotypes for GSTT1 were available , we further addressed the association between GSTT2B and GSTT1 deletion polymorphisms in HapMap populations . Consistent with the results from our Caucasian samples , LD between the two deletion polymorphisms was strong in CEU ( D′ = 0 . 841 ) , with a significant overrepresentation of alleles with the single deletion ( p = 2 . 2×10−16 ) ( Table 2 ) . In contrast , LD was less evident in JCP ( D′ = 0 . 60 ) . Association of the two deletions appears to be random in YRI ( D′ = 0 . 10 ) . In fact , data from SNP genotypes from HapMap samples in the surrounding region support our observations . There is a large haplo-block including two deletions in CEU ( Figure S2 ) . Phased haploblock analyses show that haplotypes in CEU are less diverse than in YRI ( Figure S3 ) . In order to determine whether the GSTT2B deletion can be tagged by neighboring SNPs , we also assessed LD between the deletion polymorphisms and surrounding SNPs ( Figure 5B ) ( Tables S3 , S4 , S5 , S6 , S7 , and S8 ) . HapMap SNP genotypes 500 kb to either side of deletions were obtained , and r2 between deletion polymorphisms and SNPs was calculated . LD between the GSTT2B deletion polymorphism and SNPs were observed , and SNPs with r2>0 . 7 were identified up to 35 kb of the centromeric side and 11 kb on the telomeric side of the deletion in all three populations . There were several SNPs showing strong LD ( r2>0 . 8 ) in JCP . Considering the fact that identifying SNPs showing complete LD ( r2 = 1 . 0 ) with nearby CNVs is very difficult in complex , repeat-rich regions [6] , [47] , [48] , we may conclude that the GSTT2B deletion allele is tagged by nearby SNPs and is derived from a unique ancestral allele . In contrast , LD between SNPs and the GSTT1 deletion polymorphism showed a population-specific pattern . The deleted region including GSTT1 is flanked by a pair of 466-bp direct repeat ( Figure 2A ) . The 51-kb region between direct repeat is deleted in the deletion allele of GSTT1 with only one 466-bp repeat remaining in the allele , which strongly suggests non-allelic homologous recombination ( NAHR ) as an underlying mechanism . SNPs with r2>0 . 7 were identified up to 100 kb on the centromeric side in CEU , consistent with the previous analysis [5] . In contrast , SNPs with r2>0 . 7 were less frequent and were only found within 10 kb on either side of the GSTT1 deletion in JCP . There were no SNPs with r2>0 . 7 in YRI . Therefore , the GSTT1 deletion would be found recurrently in humans , and extended LD between SNPs and GSTT1 deletion polymorphism in CEU may be the result of selection forces for the haplotype harboring GSTT1 deletion . We have observed CEU-specific LD between GSTT2B and GSTT1 deletion polymorphisms . It is currently unknown whether closely located deletion polymorphisms are often in LD . Answering this question is very difficult , because , although a number of CNVs have been identified for the HapMap samples , the breakpoints as well as the copy-numbers for each CNV have not been well defined . Each CNV region tends to cover a large genomic region that may include more than one CNV . This is the case for the deletion polymorphisms for GSTT2B and GSTT1 , in which a large single CNV region ( cnp1364 ) covers both deletion polymorphisms [6] . Recently , very high-density microarray has begun to provide the locations of CNVs with higher resolution . McCarroll et al . , have developed an extremely high-density oligonucleotide microarray ( Affymetrix SNP 6 . 0 ) and has captured CNVs in the HapMap samples with improved resolution [48] . Indeed , this approach captured GSTT2B ( cnp id 2559 ) and GSTT1 ( 2560 ) deletion polymorphisms as independent ones . Although the estimated size of the cnp 2559 is larger ( 67 . 1 kb , chromosome 22: 22 , 613 , 016–22 , 670 , 785 ) than the size from our direct sequencing of breakpoints , a genotype result for each individual is highly ( 100% ) consistent with the results from PCR assay . Therefore , the data provided by McCarroll et al . , would be valid for performing a genome-wide LD analysis . In order to determine linkage between CNVs , we first selected the CNVs using the following criteria: 1 ) we focused on the diallelic deletion polymorphisms that are denoted as 0 , 1 and 2 in the publication , which leave 361 polymorphisms; 2 ) we focused on deletion polymorphisms on autosomes and excluded 16 CNVs on sex chromosomes; and 3 ) we determined the linkage between CNVs that were on the same chromosomes . There were 1857 pairs ( combinations ) for CEU , 1734 for JPT+CHB and 2592 for YRI for linkage analysis , because some of the CNVs were only seen in one or two populations . First , we determined the number of deletion polymorphism pairs as a function of r2 and significance value ( −log10p-value ) ( Figure 6A , only for CEU ) . For both r2 and significance value , the number of pairs showed power-law distributions and the vast majority of pairs had very low r2 and −log10 ( p-value ) . This indicates that only a small number of deletion polymorphisms are in LD . However , consistent with the result from our PCR-genotyping , GSTT2B-GSTT1in CEU was in a strong LD ( r2 = 0 . 699 , −log10 ( p-value ) >15 ) ( Figure 6B , marked with red circles ) ( Tables S9 ) . Next , in order to determine whether strong LD was common for closely located CNVs , we determined the r2 and significance value as functions of physical distance ( Figure 6B ) . In fact , there were several , closely located deletion polymorphism pairs with relatively high r2 ( Tables S9 , S10 , and S11 ) . These pairs were seen mostly in CEU and CHB+JPT , but not in YRI . Overall , there was very weak association for most of the pairs , even for the ones that are closely located . Therefore , the analysis using the currently available list of deletion polymorphisms indicates that the strong LD between GSTT2B and GSTT1 in CEU seems unique and may imply the presence of selection forces in this locus .
Deletion alleles of GST genes have been known for more than a decade , long before we realized the global distribution and significant impact of CNVs on genetic variation in humans . Without knowing the major role of CNVs in genetic variation , deletion polymorphisms of GST genes might well have been accepted as common polymorphisms in humans but a rare event in the human genome . Knowing now both the prevalence of CNVs and the location of GST genes in extensive SDs , we may need to consider a more detailed genotyping of GST genes for disease association studies . Our approach using Restriction Fragment Length Polymorphisms ( RFLP ) illustrated an overall genetic diversity within the GSTT2-GSTT1 locus . Two major common variants were evident in our analysis: a GSTT2B-deletion allele and a GSTT1-deletion allele . The GSTT2B deletion extended for 37 kb and caused a nearly silenced expression of the remaining GSTT2 . Therefore , a null allele likely exists for both of the theta class of GST genes in humans . Our study revealed the high frequency of the GSTT2B deletion alleles in all three HapMap populations , particularly in the CEU population . This is in contrast to the neighboring GSTT1 deletion that is the least common in Caucasians [5] . Therefore , if there are any confounding effects of the GSTT2B deletion in the GSTT1 disease association studies , it would affect associations in Caucasians more than in other populations . Association studies between lung cancer susceptibility and GSTT1 deletion may illustrate this issue . Cigarette smoke is the main environmental risk factor for lung cancer . Cigarette smoke contains free radicals and induces oxidative damage to cellular lipids and DNA [49] . The theta class of GST exhibits glutathion peroxidase activity that protects cells from oxidative damage [50] . Recent meta-analyses show a marginal , but positive correlation between GSTT1 deletion and lung cancer for Asians , but not for Caucasians [36] , [38] , [51] . We could speculate a possible reason for this observation: high frequency of the GSTT1 homozygous deletion ( 40–60% ) and lower GSTT2B deletion in Asians may have lead to a more accurate , positive association , whereas significant associations were difficult to find in Caucasians due to low frequency of ( 10–20% ) the GSTT1 deletion and high frequency of the GSTT2B deletion . Therefore , evaluating GSTT2B deletion polymorphism may be necessary in order to accurately assess associations between theta class of GST and human diseases in the future . One of the unique features for the GSTT2B and GSTT1 deletion polymorphisms is strong LD in the CEU population . Only a small number of deletion polymorphisms are in LD among the currently defined deletion polymorphisms . However , this conclusion is preliminary , given the fact that the dataset we used has a limited coverage on CNVs , in particular on smaller ( <5 kb ) ones [48] . DNA sequence level information on CNVs [13] for a large number of individuals is necessary in order to provide an improved list of CNV pairs with strong LD . One can do this for particular pairs by developing a PCR assay for each CNV based on the sequence of breakpoints and determine if there is any strong LD between CNVs . A CNV-based assessment of LD may be useful to complement the SNP-based approach , particularly for complex loci . Because the density of reliable SNPs may be limited in complex loci , a SNP-based approach may not have enough power for reliably assessing LD . Among other pairs of deletion polymorphisms , LD was very strong in pairs of deletion polymorphisms that are located in peri- centromeric regions ( Tables S9 , S10 , and S11 ) . Low recombination rate within peri-centromeric region [52] would contribute to the strong LD . For example , both CNV 796 and 797 are located within the 80 kb peri-centromeric region of the short arm of chromosome 5 . The frequencies of deletion alleles are very high in all three populations ( 796 – 0 . 45 in CEU , 0 . 41 in JCP and 0 . 25 in YRI; 797 – 0 . 45 in CEU , 0 . 41 in JCP and 0 . 25 in YRI ) . However , in contrast to the GSTT2B-GSTT1 deletion polymorphism , LD is extremely strong in all three populations ( r2; 0 . 999 in CEU , 0 . 985 in JCP and 0 . 955 in YRI ) . Deletions would occur very early in the history of humans and have been kept in the different alleles due to the lack of recombination . This emphasizes the uniqueness of deletions , and may further support the history of selection in shaping CEU-specific LD between GSTT2B-GSTT1 deletions . A distinct pattern of LD with nearby SNPs was seen for each deletion . The GSTT2B deletion appears to be tagged by nearby SNPs in all three populations . In contrast , CEU-specific , extended LD with SNPs was seen for the GSTT1 deletion . The GSTT2 deletion polymorphism most likely occurred after human-chimpanzee divergence and the deletion allele might have been propagating within the human linage . In contrast , linkage equilibrium between the GSTT1 deletion and nearby SNPs in YRI strongly suggests that the deletion including GSTT1 have occurred recurrently in humans , possibly by NAHR between 466-bp direct repeat . In CEU , the GSTT1 deletion is almost exclusively seen in the allele that retains GSTT2B . Therefore , a potential scenario could be that the GSTT1 deletion occurred in the GSTT2B non-deletion allele and has been selected for within CEU . The GSTT1 deletion could also be selected in JCP population . However , because GSTT1 deletion might have occurred recurrently in the two major alleles , the GSTT2B deletion allele and non-deletion allele in JCP , LD with nearby SNPs would not be as evident as in CEU . We initiated this study on the assumption that the instability of large DNA-IRs may be a predisposing factor for CNVs . For example , a duplicated transgene in a 16 kb perfect palindrome ( DNA-IR ) in mice was transmitted to the progeny with very high frequency of DNA rearrangements ( >15% ) [53] . Typically , DNA rearrangements occur as a deletion of a tip and part of a DNA-IR . It was shown that nuclease processing of either a tip of hairpin structure on the lagging-strand DNA during replication resulted in two-ended DNA breaks [54] . Subsequent end joining may complete the deletion process . The GSTT2B deletion includes a part of spacer and one entire repeat , which is consistent with the proposed mechanism . However , from our results , we do not know whether rearrangements occur very frequently in this particular DNA inverted repeat . The high frequency of the GSTT2B deletion most likely comes from a unique allele propagating in humans , because these alleles likely share an identical breakpoint . This inverted repeat may not be as unstable as perfect DNA palindromes due to the presence of a 2 . 1-kb non-palindromic spacer and the sequence divergence ( 2 . 1% ) between repeats . However , it still is of note that there is one individual ( 1/44 ) who has an atypical deletion ( Figure 1 ) . Therefore , overall genotypes of the locus could be more diverse than is described here . We found severely reduced expression of the GSTT2 gene in cell lines with homozygous GSTT2B deletion , suggesting an influence on neighboring gene expression [55] . Coggan et al . , have shown previously that the GSTT2B gene has a mutation at the exon 2/intron 2 splice site that causes a premature termination at codon 196 in 28% of the Australian population . This allele was considered as a nonfunctional pseudogene ( GSTT2P ) [56] . We have also observed the same mutation in a subset of our samples from Caucasian ( 9/19 ) and African ( 2/10 ) individuals ( data not shown ) . However , regardless of the functional status ( GSTT2B or GSTT2P ) , the presence of the second GSTT2 copy and its surrounding region have potential functional influence over GSTT2 expression . Position effect may explain the reduced expression . A single functional enhancer for the pair of GSTT2 ( B ) genes could potentially reside in the deleted region . The deletion would take out the single major positive control element and leave GSTT2 inactive . Alternatively , DNA-IRs itself may have a positive synergistic effect on gene expression . Gene amplification of a drug resistance gene is very often initiated by inverted duplication [57] . Inverted duplications occur to counteract specific inhibitors by increasing copy number and gene expression . Although the unstable nature of DNA-IRs has been widely recognized , a number of large stably maintained DNA inverted repeats in the human genome [39] may also suggest an advantage of DNA-IRs in biological processes , such as gene expression and DNA replication . It is important to note that , in fibroblasts , GSTT2 is reported as a differentially expressed gene between humans and chimpanzees [58] , with a much higher level of expression in the chimpanzee . In contrast to humans , chimpanzees strictly retained both GSTT1 and GSTT2B genes in the samples tested here . Consistent with our finding , a previous study has not identified CNVs for these two genes in chimpanzees [59] , although the study was done using BAC-clone based array-CGH analysis with limited resolution ( 1 MB ) . Our results provide specific genes involved in a lineage-specific CNV , which allows us to discuss history and function of the CNV . The conserved local genomic feature ( DNA-IR ) between two species , but frequent CNVs only in humans suggests the involvement of recent selective pressure . The theta-class is considered to be the most ancestral class of cytosolic GSTs , and other classes , such as mu ( GSTM ) , alpha ( GSTA ) and pi ( GSTP ) , originated from the theta class by gene duplication [33] . Importantly , unlike alpha and mu classes that have four and five paralogues respectively , there are only two paralogues for the theta-class , GSTT1 and GSTT2 . Why then are we losing ( functionally ) one of the most conserved classes of cellular detoxification genes ? The answer may be that the theta class is dispensable due to the overlapping functions with other classes . However , there are several structural features that indicate a distinct function of the theta-class [60] , [61] . First , amino acid identity between the theta-class and other classes is very low , less than 15% in mammals . Second , the highly conserved Tyr residue , a critical residue for glutathione ( GSH ) binding in other classes , is replaced by Ser . Third , the C-terminal extension in the theta-class proteins completely buries the substrate-binding pocket and occludes most of the GSH-binding site . Accordingly , the mammalian theta class lacks the ability to bind to glutathione affinity matrices , and lacks the activity with a model substrate of GSTs , 1-chloro-2 , 4-dinitrobenzene ( CNDB ) . The least accessible substrate-binding site may indicate a much narrower range of substrates , which is in contrast to other classes that possess more open , accessible substrate binding sites for a wide range of substrates . Therefore , the compromised ability to detoxify theta-class specific substrates in humans may be related to the difference in phenotypes between two species [62] . In summary , we have characterized a high frequency deletion polymorphism of GSTT2B in a complex region of the genome . We provided a molecular approach in order to directly genotype the GSTT2B deletion , which may be useful for future disease association studies . These results confirm the unusual genetic and molecular features in the regions of segmental duplications , and the necessity of a labor-intensive approach for full understanding of the biology and disease phenotypes associated with CNVs .
Peripheral-blood cells , EBV-transformed lymphoblast cell lines , and DNA samples were collected from healthy donors [63] . Informed consent was obtained from all subjects in accordance with procedures and protocols approved by Human Subjects Protection Committee . HapMap DNA samples were obtained from the Coriell Institute ( http://www . coriell . org/ ) . Sample ID and GSTT2B genotype are listed in the Table S1 . Colorectal cancer cell lines HCT116 , Lovo , HCT15 , and HT29 were obtained from the ATCC . Ovarian cancer cell lines 2008 and 2008 ( C13 ) were gift from Dr . Toshiyasu Taniguchi ( Fred Hutchinson Cancer Research Center ) . Human primary fibroblasts ( AG16409 , AG10803 , AG09319 , AG09309 and AG09429 ) , Chimpanzee primary fibroblasts ( AG06939 , S003642 , S003649 , S006007 , S007603 ) and lymphoblastoid cell lines ( AG18354 , AG18355 , AG18356 , AG18357 , AG18358 , AG18359 , AG16618 ) were obtained from the Coriell Institute . High molecular weight genomic DNA was extracted by QIAamp DNA Blood Midi kit ( QIAGEN ) . Southern blotting was carried out as described previously [64] . Two µg of high-molecular-weight human genomic DNA were digested with a restriction enzyme , separated in 0 . 8% agarose gels . The gel was transferred to a positively charged nylon membrane ( Amersham Biosciences ) for 3 h at 75–80 mmHg pressure using the PosiBlot 30–30 pressure blotter and pressure control station ( Stratagene ) . The DNA was UV-crosslinked to the nylon membrane using the Stratalinker 1800 UV crosslinker ( Stratagene ) . To make a probe for Southern-blot analysis , we amplified human genomic DNA using PCR primers IR28-26352F , 5′-CAAGAGGCTACACAGGCAGATGTC-3′ , IR28-26980R 5′-GGGCAGAGGAACGGAAACA-3′ , and cloned the fragment by TOPO TA Cloning Kit ( Invitrogen ) . In order to genotype the GSTT2B deletion , a three primer set was designed for PCR: GSTT2B-6858 , 5′-CACTCAACACAGTAGCCTCATCGTG-3′ , GSTT2B-6857 , 5′ TGCCTCCCCTGCCTTATTTC 3′ , and GSTT2B-2B , 5′-CCTTCTGAAATGGAGCCTTTG-3′ . The reaction was performed in a duplex-PCR with a final volume of 50 µl with 1 . 0 U Taq polymerase ( GoTaq , Promega ) , 1 . 5 mM MgCl2 , 200 µM dNTPs , 10 pmol of each primer , and 50 ng of genomic DNA . The thermal cycling conditions used for amplification consisted of an initial denaturation step at 95°C for 2 min , followed by 30 cycles of denaturation at 95°C for 30 s , annealing at 60°C for 30 s , and extension at 72°C for 45 s . Duplex PCR analysis for GSTT1 was performed using the four primers as previously reported [45] . The reaction was performed in the final volume of 25 µl with 0 . 4 U of Faststart Taq polymerase , GC rich solution ( Roche , USA ) , 2 mM MgCl2 , 800 µM dNTPs , 10 pmol of each oligonucleotide primer , and 50 ng of genomic DNA . Thermal-cycling conditions consisted of an initial denaturation step at 95°C for 7 min , followed by 30 cycles of denaturation at 95°C for 30 s , annealing at 60°C for 30 s , and extension at 72°C for 60 s , and final extension at 72°C for 7 min . Automated sequencing was performed directly both on the gel-purified PCR products and the PCR product cloned into TOPO TA cloning kit ( Invitrogene , USA ) . Total RNA was extracted from cells using the RNeasy Kit ( Qiagen ) . Ten µg of total RNA was loaded onto a 0 . 9% agarose-formaldehyde gel and separated for 60 min at 100 V . RNA quality was assessed by the integrity of 28S and 18S . The gel was transferred to a positively charged nylon membrane ( Amersham Biosciences ) for 3 h at 75–80 mmHg pressure using the PosiBlot 30–30 pressure blotter and pressure control station ( Stratagene ) . The RNA was UV-crosslinked to the nylon membrane using the Stratalinker 1800 UV crosslinker ( Stratagene ) . Each membrane was probed for both GSTT2 and β-actin . Both probes were PCR amplified and cleaned using the Gel Extraction Kit ( Qiagen ) . Primers for the GSTT2 cDNA are GSTT2 cDNA 31F – 5′-AGAGCTGTTTCTTGACCTGGTGTC-3′ , GSTT2 cDNA 938R – 5′-GGTTATGTATGCTGCACCTGAGG-3′ . Each probe was labeled with [α-32P]dATP ( 3000 Ci/mmol; Perkin Elmer ) . Membranes were hybridized overnight at 65°C in modified Church Buffer ( 0 . 5 M sodium phosphate , pH 7 . 2 , 7% SDS , 10 mM EDTA ) and exposed to Kodak BioMax MS film ( Kodak ) . After probing for GSTT2 , membranes were stripped at 65°C for 2 h in 0 . 5% SDS and reprobed for β-actin to verify equal amounts of RNA in each lane . 1 to 2 µg of RNA was reverse transcribed using the Superscript First-Strand Synthesis kit ( Invitrogen ) according to the manufacturer's conditions . The real time PCR was carried out using a MiniOpticon Real Time PCR Detection System ( Bio-Rad ) . The PCR reaction contained 50 ng/µl of cDNA , 10 pmol of each of the specific primer sets for GSTT2 and RPL32 , 6 . 25 µl of iQ SYBR Green Supermix ( Bio-Rad ) master mixture ( 2× mix containing 50 U/ml iTaq DNA polymerase , 6 mM MgCl2 , SYBR Green I , dNTP mix , 20 nM fluorescein and stabilizers ) in a final reaction volume of 13 µl . All reactions were performed in triplicate . Thermalcycling conditions for GSTT2 consisted of an initial denaturation of 10 min at 95°C , 40 cycles of 15 s at 95°C denaturing and 1 min at 55°C annealing and a final extension step for 10 min at 72°C . Cumulative fluorescence was measured at the end of each of the 40 cycles . For RPL32 , thermalcycling conditions consisted of an initial 2 min at 50°C and 10 min at 95°C , followed by 40 cycles of 15 s at 95°C denaturing and 1 min at 60°C annealing . Cumulative fluorescence was measured after each of the 40 cycles . Product specific amplification was confirmed by melting curve analysis . Primers used for quantification were as follows: GSTT2 , forward , 5′-CGCTCAAGGATGGTGATTTC-3′ and reverse , 5′-AGGTACTCATGAACACGGGC-3′; RPL32 , forward , 5′-GCCAGATCTTGATGCCCAAC-3′ and reverse , 5′-CGTGCACATGAGCTGCCTAC-3′ . Relative quantification of GSTT2 gene expression was determined by construction of a relative expression calibration curve using serial dilutions of a positive control . The SNP genotypes used in this work were downloaded from HapMap Public Release #23a ( 2008-04-01 ) . SNP genotypes were obtained for 500 kb regions to either side of deletions . GSTT1 deletion genotypes for HapMap samples were obtained from the previous publication [5] . Haplotypes were determined using Phase 2 . 1 [46] . Hardy-Weinberg Equilibrium tests ( HWE test ) , pairwise-r2 value , D and D′ , Chi-square p-value for marker independence were computed using R ( genetics package ) . Association between CNVs was determined using the data by McCarroll et al . [48] . In this dataset , the locations of CNVs as well as genotypes of HapMap individuals were available . In this analysis , associations between deletion polymorphisms that are on the same autosomes were determined . For the 1857 pairs ( combinations of deletion polymorphisms ) for CEU , 1734 for JPT+CHB and 2592 for YRI , pairwise-r2 value , D and D′ , Chi-square p-value and Hardy-Weinberg Equilibrium tests ( HWE test ) were computed using R ( genetics package ) . In order to determine the distance between two deletion polymorphisms , we used a formula , ( |S1–S2|+|E1–E2| ) /2 , where S1 and S2 represent the start sites ( hg18 ) of the CNVs and E1and E2 represent the end sites of the CNVs . Chi-square p-value and r2 was plotted as a function of distance . UCSC genome Browser , http://genome . ucsc . edu/ PipMaker and MultiPipMaker , http://pipmaker . bx . psu . edu/pipmaker/ The R Project for Statistical Computing , http://www . r-project . org/ PHASE: software for haplotype reconstruction , and recombination rate estimation from population data , http://stephenslab . uchicago . edu/software . html | Common diseases such as cancer are caused by interactions between multiple genetic and environmental factors . Glutathione S-transferases ( GST ) are key enzymes in eliminating carcinogens and harmful macromolecules from cells . Based on the assumption that individuals who do not have a particular type of GST genes are susceptible to cancers , a number of studies have been conducted to find a link between GST genotypes and cancer . However such associations remain inconclusive to date . Because GST genes are clustered in repetitive , complex regions in the genome , other previously uncharacterized variations/polymorphisms may have had an impact on the data . We describe here such a genotype , a 37-kb deletion of GSTT2B gene that is found very frequently among humans . The neighboring GSTT2 gene expression is greatly impaired by the GSTT2B deletion , conferring a potentially null allele at GSTT2 . The GSTT2B deletion is non-randomly associated with another high frequency deletion of the GSTT1 gene . Therefore , a detailed characterization of this complex region of the genome revealed unexpected genetic and biological interactions of large deletion polymorphisms; this is essential to consider in future disease association studies . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"molecular",
"biology/molecular",
"evolution",
"molecular",
"biology/chromosome",
"structure",
"genetics",
"and",
"genomics/chromosome",
"biology",
"genetics",
"and",
"genomics/genetics",
"of",
"disease",
"genetics",
"and",
"genomics/population",
"genetics"
] | 2009 | Linkage Disequilibrium between Two High-Frequency Deletion Polymorphisms: Implications for Association Studies Involving the glutathione-S transferase (GST) Genes |
Human African trypanosomiasis is caused by infection with parasites of the Trypanosoma brucei species complex , and threatens over 70 million people in sub-Saharan Africa . Development of new drugs is hampered by the limitations of current rodent models , particularly for stage II infections , which occur once parasites have accessed the CNS . Bioluminescence imaging of pathogens expressing firefly luciferase ( emission maximum 562 nm ) has been adopted in a number of in vivo models of disease to monitor dissemination , drug-treatment and the role of immune responses . However , lack of sensitivity in detecting deep tissue bioluminescence at wavelengths below 600 nm has restricted the wide-spread use of in vivo imaging to investigate infections with T . brucei and other trypanosomatids . Here , we report a system that allows the detection of fewer than 100 bioluminescent T . brucei parasites in a murine model . As a reporter , we used a codon-optimised red-shifted Photinus pyralis luciferase ( PpyRE9H ) with a peak emission of 617 nm . Maximal expression was obtained following targeted integration of the gene , flanked by an upstream 5′-variant surface glycoprotein untranslated region ( UTR ) and a downstream 3′-tubulin UTR , into a T . brucei ribosomal DNA locus . Expression was stable in the absence of selective drug for at least 3 months and was not associated with detectable phenotypic changes . Parasite dissemination and drug efficacy could be monitored in real time , and brain infections were readily detectable . The level of sensitivity in vivo was significantly greater than achievable with a yellow firefly luciferase reporter . The optimised bioluminescent reporter line described here will significantly enhance the application of in vivo imaging to study stage II African trypanosomiasis in murine models . The greatly increased sensitivity provides a new framework for investigating host-parasite relationships , particularly in the context of CNS infections . It should be ideally suited to drug evaluation programmes .
African sleeping sickness , or human African trypanosomiasis ( HAT ) , currently infects around 10 , 000 people per year and threatens the lives of a further 70 million people living in 36 countries of sub-Saharan Africa [1] . Infections of domestic livestock are also a major economic problem throughout the region . HAT is caused by protozoan parasites of the Trypanosoma brucei species complex which are transmitted to mammalian hosts by the tsetse fly during a blood meal . There are two sub-species of human infective parasite . Trypanosoma brucei gambiense , which is responsible for 90% of clinical cases , causes a chronic form of the disease . T . brucei rhodesiense , which is primarily zoonotic , causes an acute form of HAT in eastern Africa . The sub-species Trypanosoma brucei brucei is non-human infectious , but is a pathogen of domestic animals . HAT has two distinct stages . The first is characterised by recurrent fever and malaise as the parasite replicates in the blood and lymph . The second begins once parasites penetrate the blood brain barrier and access the central nervous system ( CNS ) . This can occur weeks ( T . b . rhodesiense ) or years ( T . b . gambiense ) after the initial infection . Without chemotherapeutic intervention , this leads rapidly to neurological impairment , coma , and death . The drugs currently used to treat HAT can be difficult to administer , are associated with a range of side effects , and treatment failures are frequently reported [2] , [3] . The front line drugs , pentamidine and suramin , were developed over 50 years ago and are only useful against the haemo-lymphatic stage of disease . Pentamidine is the drug of choice against T . b . gambiense , while suramin is only employed to treat T . b . rhodesiense infections . Melarsoprol and eflornithine , are both used to treat the second stage of infection following CNS invasion , but both have significant disadvantages . Melarsoprol is directly associated with encephalopathic syndrome , which arises in 5–10% of cases and has a mortality rate of 50–70% [4] , [5] . Eflornithine causes less severe side effects , but it is expensive and difficult to administer , requiring four infusions of 400 mg kg−1 each day over a 7 to 14 day period , and is not used against T . b . rhodesiense . More recently , a nifurtimox and eflornithine combination therapy ( NECT ) has been introduced that is more effective against late stage T . b . gambiense infections than eflornithine monotherapy . With increasing accounts of clinical relapse and the growing likelihood of drug resistance , there is an urgent need to develop novel drugs that are safe , effective during stage II CNS disease , and easy to administer in the field . A current murine model of chronic stage II trypanosomiasis utilises the pleomorphic T . b . brucei GVR35 strain [6] . However , in vivo evaluation of compounds for trypanocidal activity using this system is both labour and animal intensive and requires 180 days post-treatment to completely assess efficacy . Non-invasive whole body bioluminescence imaging can be utilized as an alternative approach in a variety of in vivo disease models to accelerate drug discovery and development [7]–[9] . A particular advantage of bioluminescence imaging is that the detected signal can be directly correlated with pathogen load . This provides a simple , non-invasive way of quantifying the effect of therapeutic intervention in real time . The use of a stage II trypanosomiasis murine model that incorporates easily detectable bioluminescent parasites , would greatly streamline the drug testing process , particularly when assessing activity against late stage brain infections . One limitation of bioluminescence imaging results from the loss of optical resolution of parasites situated in deep tissue . This is caused by the absorption and scatter of light which decreases the detectable signal by up to ten fold per centimetre of tissue [10] . Haemoglobin is primarily responsible for this phenomenon in mammals . However , because haemoglobin absorbs light mostly in the visible blue-green spectrum , this loss of sensitivity can be reduced by using red-shifted bioluminescent reporters that emit light above 600 nm [11] , [12] . Previous reports of in vivo imaging of T . brucei infections were based on expression of Renilla luciferase , which emits light in the 480 nm range [13] , [14] , and firefly luciferase with emission at 562 nm [15] . Here , we report the development and optimisation of T . b . brucei Lister 427 and T . b . brucei GVR35 reporter cell lines that constitutively express a red-shifted mutant firefly luciferase with an emission maximum of 617 nm . These parasites are detectable in deep tissue at numbers considerably lower than previously reported and will have numerous in vivo imaging applications .
BALB/c and C57BL/6N studies were carried out under UK Home Office regulations ( Project licence PPL 70/6997 ) . The CD-1 mouse work was performed with the approval of the University of Glasgow Ethics Committee ( Project licence PPL 60/4442 ) . T . b . brucei Lister 427 , MITat 1 . 2 ( clone 221a ) bloodstream forms were cultured and maintained in vitro with HMI-9 medium [16] , supplemented with 10% foetal bovine serum ( FBS ) at 37°C in 5% CO2 . T . b . brucei GVR35 bloodstream forms were cultured and maintained in HMI-9 medium supplemented with 20% FBS , 20% serum plus ( Sigma ) and 0 . 1% methyl cellulose . Both strains were electroporated with the Amaxa Nucleofector II , using program X-001 and human T-cell nucleofector buffer , and 10 µg ( Lister 427 ) , or 30 µg ( GVR35 ) of linearised construct DNA . Transgenic bloodstream forms were selected and maintained with 1 µg ml−1 puromycin . To generate the parental integration construct pTb-AMluc ( Figure 1 ) , a 250 bp fragment derived from the T . b . brucei ribosomal DNA ( rDNA ) promoter was amplified from genomic DNA using a forward primer that introduced a 5′-SacI site and a reverse primer that introduced 3′-MluI and NotI sites ( all primers used are described in Table S1 ) . In addition , a 563 bp fragment from the rDNA non-transcribed spacer region was amplified using a forward primer that introduced a 5′-ApaI site and a reverse primer that introduced a 3′-KpnI site . The amplicons were digested with SacI/NotI and ApaI/KpnI , respectively , and ligated sequentially either side of a puromycin resistance cassette in a pBluescript backbone ( Figure 1 ) . To test 5′-untranslated regions ( UTR ) , fragments derived from T . b . brucei VSG221 ( 198 bp ) and T . b . brucei GPEET2 procyclin ( 89 bp ) genes were amplified using forward primers that introduced 5′-NotI sites and reverse primers that introduced 3′-XhoI sites . Each amplicon was NotI/XhoI digested and ligated into separate pTb-AMluc vectors . To test 3′-UTR regions , fragments from the corresponding regions of T . b . brucei VSG221 ( 198 bp ) and actin A ( 289 bp ) , were amplified using forward primers that introduced BamHI sites and reverse primers that introduced HindIII sites . Each 3′-UTR amplicon was BamHI/HindIII digested and ligated into the corresponding vectors . The red-shifted luciferase genes PpyRE9H and PpyRE-TS [12] were amplified with a 5′-forward primer that introduced an XhoI site and a reverse primer that introduced a 3′-BamHI site . Each reporter gene fragment was XhoI/BamHI digested and ligated into separate backbones , to create the final constructs ( Figure 1 ) . Prior to transfection , vectors were linearised by SacI/KpnI digestion . Luciferase assays were performed with a kit according to the manufacturer's protocol ( Promega ) . Prior to each assay , cells were grown to mid-logarithmic phase and 5×106 cells were pelleted by centrifugation . The cells were lysed by resuspension in 50 µl of CCLR buffer ( Promega ) . 10 µl of this was added to 100 µl of luciferase substrate . End point or spectral luminescent readings were taken using a SpectraMax M3 Microplate Reader ( Molecular Devices GmbH ) . Growth was monitored by counting each cloned cell line on consecutive days over a 12 day period , followed by daily dilution back to 1×105 parasites ml−1 . Construct stability was tested by culturing each clone in the presence or absence of 1 µg ml−1 puromycin and assaying luciferase activity weekly over a period of three months . Female BALB/c , CD1 and C57BL/6N mice were purchased from Charles River ( MA . , USA ) and housed in individually ventilated cages . For infection studies , mice ( 18–20 g ) were inoculated intraperitoneally ( i . p . ) with parasites in 200 µl HMI9 medium . For imaging , mice were inoculated i . p . with 200 µl D-luciferin ( 15 mg ml−1 in Mg/Ca-free Dulbecco's modified PBS ) ( Perkin Elmer ) , and 10 minutes later anaesthetised with 2 . 5% isofluorane . Light emission was recorded using Lumina or Spectrum in vivo imaging systems ( IVIS ) ( Perkin Elmer ) . Exposure times varied between 1 second and 5 minutes , depending on signal intensity . Peripheral parasitemia was scored by counting Giemsa-stained smears of tail blood . To clear peripheral parasitemia , mice were treated with a single 40 mg kg−1 oral or i . p . dose of berenil ( 1 , 3-Bis ( 4′-amidinophenyl ) triazene ) and imaged between 4 and 7 days later . Mouse brains were excised after perfusion for ex-vivo imaging . Briefly , mice were exsanguinated under terminal anaesthesia , then underwent vascular perfusion via the hepatic portal vein with 10 ml PBS . Successful perfusion was assessed by blanching of the liver . The whole brains were then excised and imaged in the Lumina IVIS . To enhance the signal and avoid desiccation , approximately 100 µl D-luciferin ( 15 mg ml−1 in Mg/Ca-free Dulbecco's modified PBS ) was pipetted onto the surface of each brain 5–10 minutes prior to imaging [15] , [17] . In accordance with local animal welfare regulations , infected mice were euthanised when they displayed immobility , hind-leg paralysis , or 20% weight loss .
In trypanosomes , protein coding genes can be transcribed by RNA polymerase I ( Pol-I ) , leading to high levels of expression [18] . We therefore designed integration vectors where the bioluminescent reporter genes would be targeted to the rDNA loci , under the control of a Pol-I dependent rDNA promoter ( Figure 1 ) . Constructs containing the red-shifted luciferase genes PpyRE-TS ( thermostable ) and its derivative PpyRE9H ( humanised codon usage ) [12] , flanked by the 5′-GPEET2 procyclin splice site and 3′-tubulin UTR sequences , were first used to transfect bloodstream form T . b . brucei Lister 427 . This is a monomorphic strain in which transformants can be generated in only 5–7 days . Parasite clones expressing PpyRE9H were found to display much higher luciferase activities than those expressing PpyRE-TS ( data not shown ) . A series of constructs was then generated to identify the combination of 5′- and 3′-UTR regions which facilitated the optimal level of PpyRE9H expression ( Figure 1 ) . Analysis of transformed clones revealed a range of luciferase activities in each case ( Figure 2A ) . Interestingly , replacement of the 3′-tubulin intergenic sequence with the corresponding sequence from the actin A gene , resulted in a general reduction of expression levels in transfected trypanosomes . The highest expression levels were found in parasites transfected with a construct ( pTb-AMluc-v ) containing the 5′-VSG and 3′-tubulin UTR pairing . We next transfected the pleomorphic T . b . brucei GVR35 strain with the pTb-AMluc-v construct . Again , we observed a wide range of expression levels ( Figure 2B ) . One clone ( VSL2 ) expressed luciferase activity that was almost 10-fold greater than the highest expressing Lister 427 clone . To establish if integration and expression of the reporter gene affected parasite growth , GVR35 and Lister 427 clones were followed in culture . There were no significant differences in the growth rate of transgenic and wild type cell lines ( Figure S1 ) . To ensure that reporter gene expression was stable over time , cell lines were sub-cultured in the absence of selective drug pressure for up to 3 months . This time span equates to more than 100 generations in the pleomorphic GVR35 cell line and 300 generations in the monomorphic 427 cell line . No significant changes in bioluminescence were detected ( Figure S2A and B ) . We assessed if there was a direct correlation between bioluminescence and cell number by conducting luciferase assays on serial dilutions of the VSL2 parasite clone ( Figure 3 ) . Linear regression analysis showed a strong positive correlation between the level of bioluminescence and the number of live cells ( R2 = >0 . 99 ) . We also sought to determine the limit of detection achievable with the Lumina IVIS . Serial dilutions of the VSL2 clone in a 96-well microtitre plate format were imaged after the addition of luciferin ( Materials and Methods ) . 100 parasites could be readily visualised ( Figure 3A ) . To determine the limit of detection in vivo , BALB/c mice were inoculated i . p . with a range of parasite loads and imaged following injection with luciferin ( Figure 4A ) . Using a maximum exposure setting , it was possible to visualise as few as 100 parasites in the intra-peritoneal space . This level of sensitivity was between two and four orders of magnitude greater than has been reported elsewhere with trypanosomatid parasites [7] , [13] , [14] , [19]–[21] . There was also a strong positive correlation between bioluminescence and parasite inoculum when the total flux was recorded from individual mice ( Figure 4B ) . This suggests the feasibility of assessing in real time the extent of parasite burden over several orders of magnitude directly from bioluminescence during the course of an infection , even when parasites are undetectable by microscopic analysis of peripheral blood . The course of infection in BALB/c mice was assessed following the inoculation i . p . with 3×104 VSL2 T . b . brucei GVR35 parasites ( Materials and Methods ) . This experimental model of chronic stage II trypanosomiasis is characterised by CNS infection and death after approximately 35 days , in the absence of drug treatment . The mice were imaged one day after infection and then at subsequent time points ( Figure 5 ) . At each time point , blood smears were also taken from the tail to quantify peripheral parasitemia . Bioluminescence imaging allowed the development of infection to be visualised throughout the entire experiment , whereas parasites only became detectable in peripheral blood 14–18 days post-infection ( Figure 5A ) . The subsequent transient decrease in bloodstream parasitemia observed after this time point presumably results from antibody-mediated killing and antigenic variation within the parasite population . During this apparent clearance of infection , as judged by microscopic examination of blood , parasitemia was readily detectable by in vivo imaging ( day 24 , Figure 5A ) . Brain infections could be tentatively identified within 7 days of inoculation , particularly when mice were imaged from the dorsal perspective . After 32 days , the mice were treated with berenil , a drug which can resolve bloodstream , but not CNS infections [22] . As expected , there was a dramatic effect , with complete clearance of parasitemia , as judged by microscopic examination of blood smears ( Figure 5A , B ) . With bioluminescence imaging however , it was clear that parasites had not been eradicated , with specific foci of infection in the cranial area . Examination of excised brains from other infected mice ( day 33 ) revealed clear evidence of wide-spread infection , with distinct , highly intense bioluminescence ( Figure 5C ) . Interestingly , bioluminescence was observable 4 days post-treatment in the region of the mouth and nose ( Fig . 5B ) . This signal could be indicative of parasite survival in the nasal associated lymphoid tissue or result from trypanosomes which had spread from the CNS via the lymphatics into the plate region . In the absence of drug treatment , the mean survival time ( Materials and Methods ) for BALB/c mice infected with the bioluminescent VSL2 clone was 34 . 4±1 . 2 days ( n = 14 mice ) . To assess the increased sensitivity achievable with the “red-shifted” luciferase , as compared to the wild type reporter gene , we carried out parallel infections of CD-1 mice . One set was infected with VSL2 T . b . brucei GVR35 trypanosomes and the second with parasites expressing a yellow firefly luciferase ( LUC2 ) [15] integrated at a rDNA locus ( Figure 6 ) . Images taken 7 days and 21 days post-infection clearly show the enhanced bioluminescent signal achievable with the “red-shifted” reporter . Similarly , when mice were then treated with berenil , residual brain infections were much more readily detectable in mice infected with the VSL-2 parasites . We also examined the sensitivity of imaging when applied to C57BL/6N mice , the most commonly used strain for the generation of transgenic lines . In this background we observed that the level of bloodstream parasitemia was lower than in the BALB/c strain . In addition , there appeared to be some quenching of the signal , presumably due to the black fur of this mouse strain . Nevertheless , the course of infection could be easily monitored with the bioluminescence system and it appeared to follow a similar pattern to that in BALB/c mice ( compare flux traces in Figure 5A and Figure 7A ) . Distinct signals appeared in the head region between day 14 and day 24 ( Figure 7 ) corresponding to the CNS infection seen in the BALB/c mice .
The major goal of the work described here was to improve the tools available for in vivo bioluminescence imaging of trypanosome infections , as an aid to drug development programmes and studies on disease pathology . We addressed two major parameters that in combination have led to a greatly increased sensitivity of detection . First , we optimised stable high level expression of the reporter gene in transgenic parasites , and second , we used a variant luciferase protein that had been modified to enhance deep tissue detection in vivo [12] . The expression vector used in this study was designed to target the multi-copy rDNA loci with the bioluminescent reporter gene under the control of a ribosomal promoter ( Figure 1 ) . Previous work had shown that this strong promoter can facilitate high level expression of protein coding genes in trypanosomes [18] . In addition , we hypothesised that in situ transcription termination signals at the targeted locus would prevent the aberrant expression of downstream genes . We chose to avoid integration at a polycistronic Pol-II locus , such as the tubulin gene array , in case perturbation of local gene expression had phenotypic consequences . There are eighteen rRNA arrays in the diploid T . brucei genome , spread over several chromosomes [23] . When transfectants were examined for luciferase activity , we found significant differences between clones derived from the same electroporation ( Figure 2 ) . This may reflect differing levels of epigenetic control , either repression or de-repression , at the targeted rDNA loci . These observations emphasise the need to examine a number of clones for luciferase activity each time a new parasite isolate is transfected with this class of construct . As post-transcriptional control of gene expression in trypanosomatids is of particular importance [24] , we also tested various sets of 5′- and 3′-UTR sequences in the context of rDNA integration , to determine the best combination for expression . As shown ( Figure 2 ) , the highest levels of expression were associated with construct pTb-AMluc-v , which contain sequences from 5′-VSG and 3′-tubulin untranslated regions . Motifs in these UTR regions of trypanosomatid genes can have a major influence on transcript processing and stability , or translational efficiency . Most in vivo bioluminescence imaging studies previously published have used luciferase reporter proteins which emit light in the 480–560 nm range [7]–[9] , [13]–[15] . However , this wavelength coincides with the region of the spectrum where light absorbance by haemoglobin is maximal , with a concomitant quenching of the signal when parasites are localised in deep tissue . Here , we used a thermostable luciferase variant from the North American firefly Photinus pyralis , which had been mutated to generate light with a peak emission wavelength of 617 nm [12] . In addition , the gene ( PpyRE9H ) had been altered to conform to human codon bias , which is similar to that of T . b . brucei [25] , and modified further to remove non-coding repeats , local hairpins and cryptic splice sites [12] . When mice were infected with transformed bloodstream forms expressing PpyRE9H in the context of the integrated pTb-AMluc-v vector , it was possible to detect as few as 100 trypanosomes when concentrated in the peritoneum , with a direct correlation between parasite load and the total bioluminescent flux ( Figure 4 ) . Parasites in brain might be less readily detected due to light absorption and scattering by the skull . The pattern of infection and pathology mirrored that obtained with non-transfected T . b . brucei GVR35 , an experimental model for stage II trypanosomiasis ( Figure 5 ) . Consistent with this , when transformed bloodstream parasites were monitored in vitro , there was no evidence of an effect on growth and reporter gene expression was stable for at least 3 months in the absence of selective pressure ( Figure S1 , S2 ) . The level of imaging sensitivity reported here significantly surpasses that which has been reported previously with trypanosomatid parasites [7] , [13] , [14] , [19]–[21] . It permits infections to be followed in real time , in a non-invasive manner , even when peripheral bloodstream parasitaemia is sub-patent . The ability to image such low numbers of trypanosomes will find widespread use for monitoring chronic CNS infections where the level of parasite burden may be low . In addition , it will allow the visualisation of tissue-specific clearance of infection after drug treatment , without the need for sacrificing mice at each stage of the process ( Figure 5B , as example ) . An additional benefit of this bioluminescent parasite line is that it is easily detectable in C57BL/6 mice ( Figure 7 ) , the background used in most transgenic experiments . With this non-invasive imaging technology , it will therefore be feasible to assess the contributions of different genes to host-parasite interactions , parasite dissemination and CNS infection , using genetically modified bioluminescent trypanosomes in combination with transgenic mouse strains . The genes encoding the “red-shifted” luciferase variants used in this study [12] should also be transferable to T . cruzi and Leishmania to enhance the applicability of imaging technology to experimental infections with these related pathogens . | Parasites of the Trypanosoma brucei species complex are the causative agents of human African trypanosomiasis . There is an urgent need for new drugs to treat this debilitating and potentially fatal infection , especially in its late stage , when parasites have entered the central nervous system . Factors which hamper drug development include the limitations of the current murine models for stage II disease . In vivo bioluminescence imaging is a non-invasive technique that can be used to monitor infections in real time and is a powerful new approach for studying drug effectiveness . However , application of this imaging technology to trypanosome infections has been restricted because of lack of sensitivity . In this paper , we have taken a major step to resolve this problem . The enhanced sensitivity in infected mice is based on the high level expression in trypanosomes of a “red-shifted” luciferase variant that greatly improves bioluminescence detection in deep tissue . The system which we have developed should be a widely applicable tool for providing new insights into the infection biology of T . brucei . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] | [] | 2013 | Highly Sensitive In Vivo Imaging of Trypanosoma brucei Expressing “Red-Shifted” Luciferase |
Cebu has been one of the most leprosy endemic areas in the Philippines . Despite the high coverage rates of multiple drug therapy ( MDT ) and high BCG-vaccine coverage in children , leprosy control authorities believe that leprosy transmission and incidence ( as evidence by continuing new case detection in both adults and children ) have not declined as expected , once leprosy had been eliminated . In response to the concerns communicated by the authorities regarding ongoing leprosy transmission in Cebu , this study aims to examine the evidence for the hypothesized ongoing transmission , both in children and adults . Furthermore , it will be assessed which groups and areas are experiencing a continuing risk of leprosy infection; this can form a starting point for more targeted approaches to leprosy control . Case records from 2000–2010 were retrospectively collected from the Leonard Wood Memorial Clinic archives , and all other clinics on the island where leprosy was treated . Between 2000 and 2010 , 3288 leprosy cases were detected . The overall five year case notification rate ( CNR ) dropped significantly from 47 . 35 ( 2001–2005 ) to 29 . 21 cases ( 2006–2010 ) per 100 . 000 population . Smaller CNRs were reported for children; however the decline in child-CNR over the same period was minimal . Furthermore , no increase in median age of notification in children or adults was found between 2000 and 2010 . Population-adjusted clustering of leprosy cases was mainly detected in urban and peri-urban areas . Although the overall CNR declined significantly , CNR seems to be rather static in lower risk populations and areas . Cases are mainly found in urban areas , however CNRs in these areas decline at a much faster rate than in the lower endemic rural areas . A similar situation was found when comparing adults and children: CNRs observed in children were lower than in adults , but further decline ( and elimination ) of these childhood CNRs was found to be difficult . Moreover , the median age of notification in children has remained stable , suggesting transmission is still on-going . It is unclear why many years of good MDT-coverage and a gradual decline in CNR have not been accompanied by evidence of reduced transmission , especially beyond a certain threshold level of case notification . We believe that a new approach to leprosy control is required to tackle transmission more directly . The most promising approach may involve chemoprophylaxis and/or immunoprophylaxis interventions , targeted at high risk ( urban ) areas and groups such as household contacts , followed by a different approach once decline in CNR starts to level off . Identified clusters and trends can form the starting point for implementing this approach .
For over a century , Cebu has been one of the most leprosy endemic areas in the Philippines . With the establishment of a centralized leprosy settlement in the island of Culion , compulsory notification of leprosy cases was introduced in 1906; ever since , a large proportion of leprosy cases have been coming from Cebu . Because of the high endemicity , another leprosy center was established in Cebu in 1930 , through the Leonard Wood Memorial Leprosy Research Center ( LWM ) . Over the next two decades , LWM conducted a number of classic epidemiological leprosy studies in the area , particularly in the municipality of Cordova . Much of our basic knowledge of the epidemiology of leprosy comes from these early population surveys [1] , [2] . To reduce the global burden of disease associated with leprosy , the World Health Organization introduced Multiple Drug Therapy ( WHO-MDT ) in 1982 . WHO-MDT is a convenient , relatively inexpensive regimen consisting of monthly rifampicin and clofazimine , and daily dapsone and clofazimine administered for 2 years ( more recently , this was reduced to one year ) in multibacillary leprosy . By 1994 , MDT was implemented worldwide and the overall prevalence of leprosy dropped dramatically [3] . In 1985 , MDT was implemented in the Philippines . After a few years , relatively good coverage rates were observed particularly in the island of Cebu [4] , [5] . The LWM clinic , which handles the majority of Cebu leprosy cases , documents through its internal reports an MDT completion rate of >95% among its diagnosed cases since 1990 [6] . The fall in global prevalence with the introduction of MDT led to the WHO campaign to eliminate leprosy as a public health problem by the year 2000 , with the assumption that once prevalence fell below the target figure of 1 case per 105 , transmission would be interrupted , leading to the gradual extinction of the disease [7] . If this process were in fact happening , health authorities would be justified in reducing the allocation of resources to leprosy control . In addition , BCG vaccination , which is believed to have some prophylactic effect against both tuberculosis and leprosy [8] , has seen increasing coverage in the past 30 years; from less than 70% in the 1980's to over 90% in the 2000's [9] . Currently the BCG coverage is estimated at 88% throughout the Philippines [9] , including the island of Cebu . Despite these measures , however , studies in the last decades have shown that the expected decline in the new case notification rate ( CNR ) and possibly incidence of leprosy has not occurred [3] , [10]–[14]; in some high endemic areas even an increase in leprosy notification rates is found [10] , [15]–[17] . Globally , approximately 250 , 000 new leprosy cases are still detected each year [17] . It is of crucial importance that the concerns of health authorities on possible ongoing transmission are studied and high risk groups and areas will be identified . Therefore , in this study we will examine whether transmission is continuing in Cebu , examining overall and group-specific CNRs and trends over 11 consecutive years ( 2000–2010 ) , as well as spatial and spatio-temporal trends . Special attention will be paid to childhood leprosy . The province of Cebu ( consisting of the island of Cebu and a few smaller islands close by ) is sub-divided into 53 municipalities , of which four comprise the greater conurbation of Cebu City and its suburbs ( these are Cebu City , Lapu-Lapu City , Talisay and Mandaue City ) . Approximately 40% of the island's population of 4 . 2 million lives in this conurbation . LWM's Skin Clinic is in Cebu City , while the residential leprosy center established in 1930 is in Mandaue City; Cordova is adjacent to Lapu-Lapu City on nearby Mactan Island . Analyzing and interpreting more than a decade's worth of data will enable us to better understand possible transmission patterns and risk profiles and will help us to prioritize and optimize leprosy control strategies .
With the approval of LWM's Institutional Ethical Review Board , the study was conducted in coordination with the Regional Health Authorities and all leprosy treatment facilities throughout the island . All data used were analyzed anonymously . The abstracted register data was digitalized by a trained database manager and anonymized before sharing the data with the data analyst/statistician . In the rare occasion a household visit had to be conducted , oral consent was obtained from the patient . This was selected as the preferred method of obtaining consent ( above written consent ) in view of the lower literacy rates in the rural areas of Cebu Island; oral consent would minimize embarrassment of the patient related to literacy . The Institutional Ethical Review Board approved this decision . There were no patients that refused consent in this study .
Between 2000 and 2010 , Cebu detected a total of 3288 leprosy cases ( Table 1 ) . A significantly declining case notification was found over the selected 11-year period; from 319 cases ( CNR: 12 . 0 per 105 population ) notified in 2000 to 204 cases ( CNR: 4 . 8 per 105 population ) in 2010 . The ARIMA ( 1 . 1 . 0 ) model shows an annual decline of 0 . 715 [0 . 474/0 . 955 , P = 0 . 002]] cases per 105 population per year ( first order Autoregressive Correlation Coefficient ( ACC ) −0 . 820 [−1 . 192/−0 . 448]; Ljung-Box χ2: P = 0 . 8592 ) . When splitting up the rural/per-urban municipalities from the four large urban areas , the data shows a drop in CNR in rural areas from 7 . 50 cases per 105 population in 2000 to 3 . 50 cases per 105 in 2010 . In urban areas these numbers are 18 . 43 and 6 . 67 respectively . Using an ARIMA ( 1 , 1 , 0 ) model an annual decrease in CNR of 0 . 380 [0 . 144/0 . 617 P = 0 . 002] cases per 105 population in the rural/peri-urban municipalities can be observed ( first order ACC: −0 . 753 [−1 . 362/−0 . 144]; Ljung-Box χ2: P = 0 . 8871 ) . The four larger cities , Cebu-City , Lapu-Lapu City , Mandaue City and Talisay showed in a similar model a much higher annual decrease of 1 . 19 [0 . 389/1 . 986 , P-value = 0 . 004] cases per 105 population ( first order ACC: −0 . 580 , [−1 . 346/0 . 185] , Ljung-Box χ2: P = 0 . 9509 ) . It appears that in urban areas the CNRs are higher , but decrease more rapidly than in the lower endemic rural areas . When looking at percentagewise differences in CNR over two periods ( Period 1 = 2001–2005 , Period 2 = 2006–2010 ) , we learn that the overall CNR reduces by 27 . 8% over period 1 , and 35 . 7% over period 2 . In the rural/peri-urban municipalities these percentages are 30 . 0 and 30 . 4 respectively and for the four larger cities 29 . 4% and 38 . 6% . The home address from 52 patients was missing . CNRs for MB and PB cases were separately analyzed . Overall the MB-CNR reduced from 9 . 45 cases per 105 population in 2000 to 4 . 23 cases per 105 population in 2010 . Using an ARIMA ( 1 , 1 , 0 ) model , an annual reduction of 0 . 509 [0 . 144/0 . 873] cases per 105 population can be observed ( first order ACC = −0 . 810 [−1 . 519/−0 . 100] , Ljung-Box χ2:P = 0 . 9262 ) . The PB-CNR reduced from 2 . 23 cases per 105 in 2000 to 0 . 56 cases per 105 in 2010 . An ARIMA ( 2 , 1 , 0 ) model was selected and showed an annual decrease of 0 . 180 ( 0 . 056/0 . 305 ) cases per 105 population ( no significant 1st and 2nd order ACCs; Ljung-Box χ2: P = 0 . 661 & 0 . 881 ) . Overall MB-CNR showed a 55 . 2% reduction over 11 years ( 18 . 0%in period 1 and 32 . 6% in period 2 ) , while PB showed a 74 . 8% reduction over eleven years ( 73 . 0% and 49 . 2% , respectively ) . The ARIMA ( 0 , 1 , 1 ) model detected a small significant increase in the proportion of patients who were MB of 1 . 05% [0 . 22–1 . 87%] per year .
Leprosy has been studied intensively in Cebu for several decades , beginning with population surveys in one municipality conducted by Doull et al . [1] , [2] during the 1930s and 40s . Further studies , including clinical trials of treatment regimens , have been conducted more recently , suggesting that basic leprosy control activities have generally been well supervised in the island . It is often assumed that good case-finding and chemotherapy with MDT , as well as a background of good coverage with BCG immunization in infants , would lead to a diminution of leprosy transmission and a decline in the incidence of leprosy . In this study , we have studied this hypothesis for Cebu Island , Philippines . We have used case notification as a proxy for incidence , taking care to ensure that case notification methods remained the same during the study period , and median age of diagnosis in children as proxy for ongoing transmission . It is an enigma that despite good MDT coverage for many years and a gradual decline in CNR , the transmission of leprosy appears to be continuing in Cebu . Furthermore , the recent analysis of data from many countries has shown that the global decline in leprosy case detection has been less than expected , despite widespread use of MDT [3] . Actual numbers of new cases reported for the last 5 years are very stable , while detection rates decline slowly due to rising population figures [9] . Our results suggest that the decline in CNR over the study period is often higher in sub-groups and areas with a higher ‘start CNR’ , and seems to decline very slowly if the CNR in 2000 was already relatively low . This could indicate a threshold , after which it becomes more difficult to lower CNR . This observation has been very consistent throughout our results; the decline of CNR was much faster for high endemic urban areas than in lower endemic rural areas ( and those outside clustered leprosy areas ) , for men ( high CNR ) it declined much faster than for women ( lower CNR ) , and for adults ( in different age groups ) the decline was much faster than in children ( with lower CNRs ) . This suggestion is however not in line with the reduction in CNR of PB versus MB patients , where we observed that the lower CNR of PB-cases drops slightly faster than the CNR of MB patients . This could be a result of the fact that only passive case finding was conducted in the period under study , following active case finding activities in 1999 . During the decade 2000–2010 , a total of 3288 leprosy cases were detected in the island of Cebu , with a declining trend . Despite these promising reductions in overall CNR over the last 11 years , notification rates in children are declining much slower , suggesting transmission is still ongoing . The CNR in children under 15 years of age has remained quite stable at around 2 per 100 , 000 population ( Figure 1 ) and the median age of children at diagnosis ( approximately 11 years ) has not changed significantly over the decade Figure 2 . If transmission had been greatly reduced over the last two decades , as many believed would occur as a result of the elimination campaign , one may expect a reduced number of child cases , especially in the youngest age group and rising average or median age at diagnosis , amongst those children who do get leprosy . It should be noted that the MB/PB ratio should be taken into account , when comparing median age of diagnosis , as a general tendency of earlier/later diagnosis ( as result of awareness campaigns , changes in health care fees etc . ) could distort age analysis based on CNR . In his comprehensive review , in 1985 , of age-specific leprosy data in situations of gradually declining incidence , Irgens [14] demonstrates that in both leprosy and tuberculosis , two apparently opposing trends can be identified as the overall incidence declines , which may make interpretation more complicated . The evidence suggests that when transmission still occurs in a population , infection tends to occur at quite a young age , but because of the variable and often very long incubation period , the onset of disease may be at any age . Thus , on the one hand , in cross-sectional studies ( looking at all the new cases in any particular year ) , older people will have been infected at a time of higher rates of transmission , when they were children , and will therefore have a higher lifetime risk of developing disease; they will be over-represented amongst new cases , so the age of onset of active disease will appear to be gradually increasing , as less and less disease is diagnosed in young people who have a lower lifetime risk of disease . On the other hand , if one examines disease in any particular cohort ( for example , everyone born in 1950 , or 1999 , etc . ) , a different pattern will be observed , with a maximum incidence at age 15–25 years; the overall incidence in the later cohorts will be less , but the general pattern will remain the same for each cohort examined [14] . In Irgens' editorial [14] , it is worth noting in greater detail the reported trend of CNR in children in Norway , between 1851 and 1920 ( his figure 1 ) , which declined from around 15 per 105 to 0 . 1 per 105 population . The rate declines to just under 2 per 105 population in the period 1881–1900 , and then to 0 . 1 per 105 for the period 1901–1920 . The periods of review in our study are shorter ( two 5 year periods ) , but it seems surprising that only a very small reduction in childhood CNRs was detected , if leprosy really is being eliminated Surprisingly , although the prevalence rate of leprosy was much higher in the 1930s , Doull et al [1] reported that childhood leprosy occurred in the different sub-groups in very similar proportions to those reported here: 3 ( 5% ) in the age-group <5 years; 22 ( 38% ) aged 5–9; 33 ( 57% ) aged 10–14 , amongst a total of 58 cases under 15 years of age; in the current study the proportions are 6% , 39% and 55% respectively . In the COLEP study in northern Bangladesh [19] , over the first two years of follow-up , under 5s were excluded from the study , but there were 5 ( 29% ) incident cases aged 5–9 , and 12 ( 71% ) aged 10–14 . Most studies do not break down the <15 age group into smaller sub-groups , so it is difficult to speculate further on the significance of these findings . The close similarity in the proportions of children of different ages affected now , as compared with the proportions in the 1930s , suggest that some aspects of leprosy transmission to the next generation may have changed less than we think . The data also show , however , that in specific areas where the greatest reduction in overall CNR has taken place , particularly in the Cebu City area , this seems to be associated with a lowering child CNR ( although not significant in this study ) . One possible explanation may be rising living standards in the urban metropolis of Cebu City which may have a greater effect on transmission than other measures . The corollary of this is that in the other parts of the island , transmission of leprosy to the next generation appears to be continuing unchecked . Children continue to represent slightly over 12% of cases , a long way from WHO's goal of <3% , suggesting continuing transmission of leprosy in the island . Another possibility in relation to continuing transmission is the development of drug resistant leprosy . Dapsone resistance developed in many places in the 1960s and 70s , leading to the introduction of MDT in 1982 . Since that time , drug resistance has not been a problem and in recent years this has been confirmed by a drug resistance surveillance program set up by WHO in 2006 [20] . LWM is the sentinel site in the Philippines for surveillance of drug resistance in leprosy and has contributed to the development of field-friendly tests for rifampicin resistance [21] . These studies have failed to show any rifampicin resistance in leprosy in the Philippines at the present time . This study was subject to a number of limitations . First of all , the study is retrospective and no real-time verification could be made . The study uses data from a period of only passive case finding: the CNRs might therefore not be representative for incidence , and trends might have been over or under estimated . The Island of Cebu is however relatively well covered by leprosy clinics and many satellite clinics , making it accessible for the Cebu population , minimizing the gap between CNR and real-time incidence . The case numbers in some areas are small and this study evaluates a relatively short period , conclusions might therefore be somewhat tentative . However , despite these limitations , the Cebu community is stable and of manageable size; therefore the data is believed to be relatively reliable . LWM works in close collaboration with government clinics and health authorities , which suggests good reliability of the data . In conclusion , our study shows that leprosy transmission is still very active in the island of Cebu , despite good coverage with MDT and BCG in recent decades . It seems that especially in groups and areas with lower leprosy rates , such as children and people in rural areas , further reduction of CNR ( and eventually elimination ) is difficult to establish . We believe that a new approach to leprosy control is required to tackle the issue of transmission more directly; the most promising approach is likely to involve interventions , such as chemoprophylaxis and/or immunoprophylaxis , targeted at high risk groups , such as household contacts and high risk areas , with a subsequent specific approach once declines in CNRs start to level off . In Cebu , the cluster analysis in Figure 6 ( looking at both trends in space and over time ) could be used to further target new interventions . | The island of Cebu is one of the most leprosy-endemic areas in the Philippines . Multiple drug therapy ( MDT ) , improved BCG-vaccine coverage and active case finding have significantly lowered the adult case notification rates ( CNRs ) , but the CNR in children ( which is a proxy indicator of ongoing transmission ) seems to be more static over the last decade ( 2000–2010 ) . The long incubation period of leprosy , hampers determination of time of infection , however one would expect the median age of notification in children to increase when transmission decreases , as younger subjects would have a lower risk of infection: in this study no significant changes in median age were found between 2000–2010 . Furthermore , leprosy seems to be mainly confined to urban areas , where nonetheless the measured decrease in CNR is much larger than in less endemic rural areas . It is unclear why the significant decline in CNR have not been accompanied by evidence of reduced transmission and why CNR seem to level off beyond a certain threshold level . We believe that more targeted approaches ( e . g . focused on household contacts in urban areas ) involving chemoprophylaxis and/or immunoprophylaxis , followed by a specific approach for lower CNRs , are required to tackle leprosy more directly . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] | [] | 2013 | A Retrospective Study of the Epidemiology of Leprosy in Cebu: An Eleven-Year Profile |
The calcium calmodulin-dependent protein kinase II ( CaMKII ) is a dodecameric holoenzyme important for encoding memory . Its activation , triggered by binding of calcium-calmodulin , persists autonomously after calmodulin dissociation . One ( receiver ) kinase captures and subsequently phosphorylates the regulatory domain peptide of a donor kinase forming a chained dimer as the first stage of autonomous activation . Protein dynamics simulations examined the conformational changes triggered by dimer formation and phosphorylation , aimed to provide a molecular rationale for human mutations that result in learning disabilities . Ensembles generated from X-ray crystal structures were characterized by network centrality and community analysis . Mutual information related collective motions to local fragment dynamics encoded with a structural alphabet . Implicit solvent tCONCOORD conformational ensembles revealed the dynamic architecture of inactive kinase domains was co-opted in the activated dimer but the network hub shifted from the nucleotide binding cleft to the captured peptide . Explicit solvent molecular dynamics ( MD ) showed nucleotide and substrate binding determinants formed coupled nodes in long-range signal relays between regulatory peptides in the dimer . Strain in the extended captured peptide was balanced by reduced flexibility of the receiver kinase C-lobe core . The relays were organized around a hydrophobic patch between the captured peptide and a key binding helix . The human mutations aligned along the relays . Thus , these mutations could disrupt the allosteric network alternatively , or in addition , to altered binding affinities . Non-binding protein sectors distant from the binding sites mediated the allosteric signalling; providing possible targets for inhibitor design . Phosphorylation of the peptide modulated the dielectric of its binding pocket to strengthen the patch , non-binding sectors , domain interface and temporal correlations between parallel relays . These results provide the molecular details underlying the reported positive kinase cooperativity to enrich the discussion on how autonomous activation by phosphorylation leads to long-term behavioural effects .
The calcium calmodulin-dependent protein kinase ( CaMKII ) is a multifunctional , multi-subunit eukaryotic protein kinase ( EPK ) . It has key roles in calcium regulation of neuronal and cardiovascular physiology [1–3] . EPKs mediate reactions whose malfunction promotes disease across a broad physiologic spectrum . The canonical EPK has a distinctive bi-lobed structure that exploits diverse strategies to achieve allosteric regulation [4] . CaMKII has a canonical kinase domain ( KD ) tethered via a linker to an equally well-conserved association domain ( AD ) that forms a central hub of the dodecameric holoenzyme with two hexamer rings that stack with mirror symmetry . Linker diversity generates isoforms and splice variants . The kinase is activated by rises in cellular calcium that enable calcium-calmodulin ( Ca2+/CaM ) to bind to and displace an autoinhibitory regulatory domain . The autoinhibitory domain occludes interaction of CaMKII with anchoring proteins , such as NMDA receptor subunit GluN2B [5 , 6] . More subunits are activated at an increasing frequency of calcium pulse trains . The tuning frequency depends on the switching kinetics between the holoenzyme open and closed states [7] . In the closed state , the substrate binding surface is occluded by an auto-inhibitory segment composed of an N-terminal segment ( R1 ) and C-terminal α-helices ( R2 , R3 ) and ATP affinity is low [8] . R1 contains the primary auto-phosphorylation site ( T286 in α subunit , T287 in others ) , as well as residues for O-GlcNAC modification ( S279 ) and oxidation ( M281 , M282 , C290 ( δ isoform ) ) . R2 has a CaM recognition motif . Ca2+/CaM binding displaces the regulatory segment to switch CaMKII to the open state , followed by segment capture and T286 autophosphorylation by an adjacent , open ( activated ) subunit [8 , 9] in the holoenzyme . R3 has additional threonine residues ( T305 , T306 ) that are primarily phosphorylated when CaM dissociates from a Ca2+-independent ( autonomous ) kinase [10 , 11] to counteract T286 phosphorylation [12] . X-ray crystal structures show that R1 and its binding determinants in the KD core adopt different conformations . Electron spin resonance ( ESR ) reports that R1 becomes unstructured upon T286 phosphorylation , while R2 and R3 are disordered in solution [13] . Persistent CaMKII activity autonomous of Ca2+/CaM underlies conversion of transient synaptic stimulation to long-term potentiation ( LTP ) ; a fundamental problem of neuronal CaMKII biochemistry that underlies certain forms of learning and memory [6 , 14–17] . Recent studies have identified more than a dozen human mutations in CaMKII that result in various degrees of learning disabilities [18 , 19] . These mutations , when mapped onto structure , localize largely with R1 or mutations that alter KD interactions with substrates rather than R2 . The previously described binding mutations form S ( substrate binding ) and T ( Thr286 docking ) sites [20 , 21] ) . The substrate binding determinants overlap with the R1 binding groove that can be partitioned into A , B and C sites [8 , 22] . The T-site is eliminated upon release of the autoinhibitory segment by rotation of helix αD whose movement helps form the B site [23] . An outstanding issue is whether the human mutations act independently or as part of a collective network to disrupt allosteric communication responsible for frequency tuning [24] . The conventional form of autonomous activity follows T286 trans-auto-phosphorylation , which maintains activity after dissociation of Ca2+/CaM . Structures of CaMKII complexes with R1 regulatory segments of one subunit ( “donor” ) captured by the adjacent subunit ( “receiver” ) upon Ca2+/CaM binding , give important snapshots into trans-phosphorylation [8 , 9] . A hydrophobic clamp of helix αD residues within the receiver KD captures the R1 in an interaction that propagates throughout the crystal lattice . Here , we used tCONCOORD , based on stochastic distance constraints [25] , and explicit solvent molecular dynamics ( MD ) as complementary methods to address the mechanistic basis underlying the medical phenotypes of the human mutations . tCONCOORD ensembles from KD structures described collective motions and conformational coupling between the nucleotide and substrate binding sites . MD examined fine-grained architectural modulations due to primary site phosphorylation on the long-range allosteric network to establish an analytical framework for the coupling within and between subunits . We characterized the long-range allosteric relays in a KD dimer extracted in silico from the crystal lattice as well as KD monomer structures . Two strategies analyzed the network architecture based on mutual information . First , the centrality profile of the complete network of local correlated motions was determined to identify relays of the most strongly coupled fragments . Second , the protein was partitioned into contiguous sectors , with coupled dynamics , that act as semi-rigid bodies–henceforth defined as “communities” [26] . Community size and cross-talk tracked the transition between the auto-inhibited and phosphorylated states . We show that the captured R1 acts as a spring , rather than a “grappling hook” [8] to couple nucleotide and substrate binding sites [27 , 28] . The capture freezes out C-lobe core motions to generate long-distance signal relays between R1 segments of adjacent subunits . The human mutations align along the relays . T286 phosphorylation strengthens the relays to suggest how it might facilitate the inter-subunit conformational spread and modulate affinities at distant sites for calmodulin [29] and cytoskeletal actin [30] . Notably , we identify a hitherto uncharacterized sector to guide the design of allosteric inhibitors . The simulations should complement experimental ESR [13] and FRET measurements [31 , 32] on activated CaMKII dimers for dissection of the activation mechanism .
Fig 1A provides a road-map of the architectural elements investigated in this study . The R1 with T286 binds to a groove within the adjacent receiver KD C-lobe that overlaps with the receiver’s S-site and has been sub-divided into three pockets A , B and C [8] in the nematode C . elegans KD dimer ( PDB ID:3KK8 ) ) . The A-site forms a canonical substrate docking interaction with the ATP binding cleft that contains the DFG motif ( D156FG158 ) ; the B-site forms a hydrophobic patch with helix αD; the C-site has acidic residues for salt-bridges with R1 . Residue positions with mutations in the human homolog that result in learning disabilities as well as S-site and T-site mutations are mapped onto the donor KD . The human mutations localized to R1 ( I272 , H282 , R284 , T286 ) , or overlapped with , or were adjacent to the S and T-sites ( F98S , E109 , P138 ) . However , a significant fraction mapped elsewhere . Variable residue positions , identified from the multiple sequence alignment ( MSA ) of > 500 KD sequences , localized to surface accessible loops when mapped onto 3KK8 . The MSA shows strong sequence conservation between species and minimal differences between isoforms within species . The phylogenetic tree indicates that nematode and human sequences are among the most distantly related ( S1A and S1B Fig ) . Secondary structure predicted from the MSA for the C . elegans , rat and human sequences matched that observed in the crystal structure . The predictions back ESR evidence for R2 and R3 disorder [13] . The human mutations localized to mainly conserved residue positions in both rigid and flexible segments of the KD C-lobe and R1 . The flexibility profile recorded by the root mean square fluctuations ( RMSF ) of residue positions computed from the tCONCOORD ensembles of 3KK8 dimers ( Fig 1B ) is largely consistent with the B-factor values in the crystal structure with differences due , in part , from crystal contacts . ( S1C Fig ) . There are also differences between donor and receiver domains that are considered below in detail . We analyzed local protein dynamics to decipher allosteric signal propagation [33–35] . The 3KK8 dimer was extracted from the crystal structure electron density with the R1 of the donor KD captured ( “bound” ) by the adjacent receiver; while the R1 of the receiver KD was unconstrained ( “free” ) , immersed in the solvent . Three , separate , explicit solvent MD replica runs of the 1 . 7-angstrom resolution structure of the phosphorylated 3KK8 dimer and its non-phosphorylated derivative was performed . Four-residue fragments were encoded by the structural alphabet ( SA ) [36] in structures within the conformational ensembles and trajectories . In the network , the nodes were the fragments ( n = 560 ) , while edges were the normalized mutual information ( nMI ) filtered set of possible correlations ( n* ( n-1 ) /2 ) = 156520 ) ( Methods ) . The map reflected the contribution of individual fragments to the dynamics and is optimal for detection of network nodes ( Fig 2A ) . The T and S sites emerged as prominent nodes in addition to the captured R1 However , the network architecture is too dense to visualize the spatial extent or interaction strength of the connections formed by individual nodes . Communities ( n > 3 ) , each represented as a node , greatly simplify network visualization of the connectivity by encoding concerted domain motions intermediate between global motions and residue-level RMSF fluctuations . Community membership and interaction strength were computed based on spectral decomposition [37 , 38] . Our methodology reproduced the published community map of the PKA catalytic subunit ( PDB ID: 1CMK ) [26] and showed that the CaMKII KD had the similar dynamic architecture to PKA with the ATP binding site forming the community hub ( Fig 2B , S2 Fig ) . Three major communities ( A , B , C ) converged at the ATP site . The N-lobe community A included the ATP binding cleft and part of the A-site . The C-lobe was split along its centre into community B with B-site helix αD and community C with the C-site acid residue pocket . Donor KD communities B and C coalesced with dimer formation ( B’ ) , while one community ( A’ ) accounted for the receiver KD . The top couplings from the global network scored on nMI were superimposed on the community architecture ( Fig 3 ) . The captured R1 replaced the ATP binding site as the community hub . The couplings congregated around it to link the donor A site with the receiver B and C sites . We compared the 3KK8 KD with other CaMKII KD structural states in the Protein Data Bank ( PDB ) to first correlate community dynamics with monitors of kinase activity . Activation involves inward motion of the DFG side-chains for nucleotide binding and inward tilt of helix αD for contact with substrate peptides in many diverse EPKs [39] . The states could be grouped into three categories–open , inhibitor-bound and auto-inhibited . tCONCOORD conformational ensembles were generated . The dominant conformations for each ensemble were identified from cluster analysis and aligned ( Fig 4A ) . The two open activated states ( 3KK8 , 2WEL ) and the inhibitor-bound form ( 3KL8 ) had inwardly oriented DFG motifs and αD helices . In contrast , the auto-inhibited states ( 2BDW , 2VN9 , 3SOA ) had outwardly rotated helix αDs . While 3SOA has an outwardly oriented DFG motif , the loops for 2BDW and 3SOA had inward orientation albeit at a position distinct from that assumed by activated states . We next catalogued the community membership of the DFG motif ( I ) , fragment L97FEDIVAR104 from B-site helix αD ( II ) and fragment P235EWD238 from the C-site acid pocket ( III ) ( Fig 4B–4D ) . The open forms ( Fig 4B ) differed in their community architecture owing to the bound calmodulin . The calmodulin fragmented community B , with R1 dynamics uncoupled from the C-lobe . In both forms , fragments I and II had multiple memberships in communities A and B . The CaMKIINtide inhibitor peptide associated with sites A , B and C similarly to the captured R1 [8] ( Fig 4C ) . Accordingly , community interactions are substantially increased , even though the communities did not coalesce . Community interactions were weak in the auto-inhibited states ( human δ , human αfull-length subunit , and a KD extracted from the nematode coiled-coil dimer ) ( Fig 4D ) . Community A extended into the C-lobe when AD contacts limited the relative motions of the N and C-lobes to each other ( 3SOA ) . Fragment II connected the dominant communities as did fragment I in the auto-inhibited human δ , as well as the human α KD . In conclusion , the coupled ATP binding DFG motif ( Fragment I ) and helix αD ( Fragment II ) form a universal hub in activated and inactivated CaMKII KD states based on their multiple community membership . The hub is robust to variations in community interaction strength . Inward movement of helix αD is diagnostic for activation , but that of the DFG motif not necessarily so . We next characterized collective motions , extracted by principal component analysis ( PCA ) to better understand the monomer to dimer transition . The PCA projected the principal eigenvectors from different replica runs onto common principal axes , as outlined in Methods . Eigenvectors were also compared [40] for the principal motions ( PCs ) ( S3 Fig ) . The overlap was comparable to that reported for similar size assemblies [40] consistent with confinement of the essential modes within a reproducible subspace [41] . PC1PC2 plots compared the amplitudes of the dimer motions obtained by MD and tCONCOORD ( Fig 5A ) . The subspace positional fluctuations along the first two PCs accounted for >70% of the total motion . The essential dynamics represented by the first ten PCs primarily delineated relative motions of N and C lobes ( S3 Fig ) . The initial drift could result from Brownian motion and/or shallow local minima in the 3D energy landscape [42] . The PC1PC2 conformational space sampled during each MD run was typically greater 2/3 the space sampled by the tCONCOORD ensemble . The space sampled by the combined MD trajectories had considerable overlap ( > 90% ) with the ensemble . Motions of the subunit KD cores and R1 segments were compared to evaluate the extent to which captured R1 motions determined dimer flexibility . The mechanical hinge and shear surface responsible for the PC motions formed community interfaces , with the ATP binding cleft as the hub in the monomer ( S1 Movie ) . The captured R1 was revealed to be the central hinge or hub for the dimer PC motions ( S2 Movie ) . The PCA shows that the captured R1 acts as an allosteric effector to freeze-out the motions of the kinase core . Simulations on other CaMKII KD structures revealed the principal dimer motions were three-fold greater relative to monomer KD core motions . PC1PC2 spread of the receiver KD core was reduced ( > 60% ) relative to the spread of donor KD core and , indeed , KD cores from the other structures ( Fig 5B ) . This result is consistent with the differences in computed B-factor values for the donor and receiver KDs ( S1C Fig ) . The PC1PC2 plots for the R1 segment ensembles formed two distinct groups ( Fig 5C ) . The group with smaller spread represented auto-inhibited structures . The cluster with larger spread ( 4x ) consisted of the captured R1 and free R1 segments . The PC1PC2 spread of the captured R1 was reduced by 50% relative to the free R1 segments . The scenario most simply consistent with these results is of a 3-state transition between autoinhibition and activation . R1 is immobile docked as pseudo-substrate ( “auto-inhibited” state ) , mobile when free in solution ( “open” state ) and constrained upon subsequent capture ( “activated” state ) . The mobility of the KD core does not change when R1 is displaced but it is notably reduced upon R1 capture . Therefore , R1 freezes KD core collective motions more effectively when captured than when docked as pseudo-substrate , but with a compensatory increase in its own flexibility . The normalized mutual information ( nMI ) provides an information theoretic measure of the coupling . This is consistent to the Shannon entropy H ( X ) for pairs of fragments albeit incomplete because of finite sampling [43] . We first compared the end-to-end distance distributions of the captured relative to the free R1 to assess constraint ( Fig 6A , S3 Movie , S4 Movie ) . The captured R1 was constrained to a highly-extended subset of conformations . Secondary structure analysis based on the SA ( Fig 6B ) revealed the extended configuration resulted from melting of two donor R1 α-helical segments N272RERV and D284VDCL; while the intervening alanine-rich V277ASAI sequence largely retained α-helical character . The result was in line with mean residue helical propensity [44] of the fragments ( -0 . 18 ( V555ASAI ) > -0 . 29 ( N551RERV ) and 0 . 49 ( D563VDCL ) kcal / mol ) . Our results are in excellent agreement with ESR experiments where spin labels were introduced at single residue positions throughout R1 in an inactivated monomeric C . elegans KD . The labels reported correlation times consistent with loop and α-helical configurations for residues S278AIHRQ and D284VDCL respectively [13] . We next focussed on the C-lobe to evaluate the compensatory decrease in flexibility of the core with greater precision ( Fig 6C ) . Comparison of the RMSF profile of the dimer C-lobes shows that reduced flexibility of the B site helix αD and the interfacial surface of the receiver lobe are responsible for the overall decrease reported by PCA ( Fig 5B ) . The solvent exposed area and electrostatic properties of the phosphorylation site were calculated for the most populated conformations extracted by cluster analysis . Dimer conformations from each set of MD replicas , as well in the tCONCOORD ensemble with the phosphomimic mutation ( TPO286D ) [1] were analysed . Dimer contacts were localized to the C-lobes . Contact between R1 of the donor subunit with the receiver subunit core upon capture accounted for much of the stabilization due to solvation . The total buried surface area due to subunit capture was 2319±42 A2 . The contribution to this value due to R1 capture was 1765±81 A2 ( ~ 75% ) . The free-energy difference due to the decrease from solvation was twice as great for the phosphorylated ( -7 . 4 kcal/mole ) versus the non-phosphorylated ( -4 . 2 kcal/mole ) form . Computed Poisson-Boltzmann electrostatic fields are shown in ( Fig 7A ) . The R1 binding pocket was more acidic and the dimer interface more polar when T286 was phosphorylated ( TPO286 ) or mutated ( D286 ) . Polar residues with large ( > 0 . 5 ) pK shifts between the T286 and TPO286 forms were identified by comparison of the dominant cluster configurations ( Fig 7B ) . These residues localized to the C-lobes around the R1 binding pocket and the subunit interface ( Fig 7C ) . Phosphorylation shifted the R1 binding pocket surface charge to negative values , while its interior became more polar . The measured pK shifts accounted for the more rigid interface . Community membership mapped onto structure revealed the captured R1 segment and its binding pocket formed a central community ( B’ ) distinct from A’ . B’ had strong interactions with donor KD communities A and B . The adjacent R1 segments were linked by long-range couplings between B’ and B . The nucleotide binding pocket and adjacent fragments of its donor KD also linked to the captured R1 ( Fig 8A ) . Superposition of the network centrality profile onto the structure delineated a signal relay from the donor A-site to the captured R1 via helix αD ( B-site ) in the receiver to its R1 ( Fig 8B , S5 Movie ) . R1-R1 communication was increased upon phosphorylation ( T -> TPO ) due to the creation of strong couplings ( nMI > 0 . 09 ) between captured R1 , the donor A and receiver B and C sites ( Fig 8C ) . These couplings represented the tail of the nMI distribution between the 2σ -3σ significance level ( S4 Fig ) . The dominant signal relay was remarkably similar to that formed by the top couplings from the tCONCOORD TPO286D ensemble ( Fig 3 ) . Superposition of the TPO dimer centrality profile highlighted the relay was localized to the captured R1 , donor A , C and receiver B sites ( Fig 8D ) . Parallel network signal processing in the networks was quantified with community size ( number of residues ) and interaction strength represented , as in Fig 2B , by community graphs ( Fig 8E and 8F ) . Phosphorylation strengthens B -> B’ interactions at the expense of A’ interactions . Two small communities mediated A -> B’ and B -> B’ signal relays . These were generated upon R1 capture and strengthened upon phosphorylation . Most human mutations causing learning disabilities map within or close to sectors that form part of the R1-R1 relay network . A sector common to both communities comprised a loop with a proline cluster ( P212 , P213 , P235 ) between C-lobe α-helices 9 and 10 that is well conserved , polar in character and at the opposite face to the R1 binding site ( Fig 8G , S6 Movie ) . The P212 mutation has an abnormal electrophysiological phenotype [18] , though it does not influence binding interactions . The loop is an attractive target for the design of positive allosteric modulators based on these characteristics . The proline mutations could also regulate interface flexibility in addition to communication between RI and the intermediate relay communities ( Fig 7G ) , Transmission of information between subunits encoded by the phosphorylation-dependent couplings was analysed to understand signal relay kinetics . The core network of long-range allosteric couplings ( n = 115 , nMI > 0 . 09 ) linked the bound and free R1 to sites A , B and C ( Fig 9A ) . A patch of three hydrophobic residues that attach R1 to B site helix αD ( I101 , A280” , I281” ) formed the central node . This node is less compact in the native ( T286 ) relative to the phosphorylated ( TPO286 ) dimer . C . elegans CaMKII mutations I101D and I281D in the hydrophobic patch that replace the non-polar isoleucines with charged aspartates abolish cooperativity ( Hill coefficient 1 . 4 -> 1 . 0 ) in decoy dimers [8] in agreement with our simulations . Similar measurements are not available for either TPO286D and TPO286A , but extensive measurements of multiple parameters in vitro and cell culture have established that these mutations mimic the phosphorylated and dephosphorylated forms of the enzyme respectively [1] . We generated tCONCOORD ensembles of engineered mutations in silico to compare the eigenvector centrality profile of a double I101D . I281D mutant with those of TPO286D or TPO286A . Comparison of the profiles implied a better match between I101D . I281D and TPO286A , with differences from TPO286D localized to two segments ( S5 Fig ) . This suggests that the TPO286A phenotype may result from loss of cooperativity; while the TPO286D phenotype has increased cooperativity . Time series of the local , SA-encoded structural transitions for residue fragments that constitute this phosphorylation-dependent network are plotted for one replica run ( Fig 9B ) . The most frequent transitions were short ( < 10 ns ) and sampled a restricted range of dihedral angles . Dihedral angle jumps scale with the separation between SA letters . Large jumps were rare but persisted for longer times ( > 100 ns ) when they occurred . Two examples are shown . The large jump for fragment V163 loop to α-helix is around the middle ( ~150 ns ) of the time series . There is a smaller jump for fragment E274 between loop ( β-sheet to α-helix ) configurations at a similar time . The cross-correlation ( CCF ) between the nMI time-series for the long-range ( > 12 angstrom ) couplings in the direct ( A279-E274 ) and indirect ( V163-E274 ) R1-R1 pathways in the phosphorylated ( TPO286 ) dimer network showed two peaks consistent with the short duration of fragment structural transitions ( 1–2 ns τ1/2 ‘s ) ( Fig 9C ) . The first peak ( Δt < 1 ns ) presumably represents couplings associated with the direct pathway . The second peak ( Δt ~ 10 ns ) represents the lag associated with the first stage of the indirect , two-stage pathway . The CCF for the end-to-end fluctuations of the two R1 segments in the phosphorylated dimer reported a weak correlation ( amplitude = 0 . 35 , correlation time τ = 15 ns ) . The correlation was not significant for the non-phosphorylated dimer .
Dynamic couplings consider both population shifts and propagation timescales of conformational ensembles [47] . Thus , they provide a more refined readout for mechanistic analysis than the architecture of the protein fold [48] . The auto-inhibited CaMKII form is completely inactive since the ATP binding site is not in an optimal conformation and the substrate binding site is occluded by the regulatory segment . The multiple community membership of the DFG and helix αD sites drove global adaptations of the network architecture coupling nucleotide and substrate binding . Reduced coupling between the communities , most marked for the coiled-coil dimer ( BDW ) structure , characterized the inactivated state . R1 capture results in reduced flexibility of the receiver C-lobe core; reported by RMSF , PCA , cluster and community network analysis . The monomer network is co-opted in the chained dimer to generate a network that spans both KDs . A hydrophobic patch between the R1 central residues and helix αD orchestrated R1donor-> R1receiver signal relays; most dominantly from the patch directly to R1receiver N-terminal segment and indirectly via the receiver DFG motif . Interfacial interactions linked the donor DFG motif to the patch . The prominent fragment couplings lasted a few to a hundred nanoseconds . Most were confined to short ( ~ 10 ns ) transitions over a small conformational range . Large conformational transitions were infrequent but persisted for longer ( >100 ns ) times . Transitions between phosphorylated and dephosphorylated conformations completed within a hundred nanoseconds . The temporal behaviour is consistent with a complex free-energy landscape with multiple pathways [49] . We suggest the reduced flexibility of the receiver KD C-lobe upon R1 capture will influence the displaced receiver R1 to explore and bind to adjacent “open” subunits with floppy cores rather than re-associate with its own frozen , occupied core . The outcome is a self-reinforcing network that can be serially transmitted across the ring of KD C-lobes . Our survey of CaMKII KD monomer structures revealed the nucleotide binding DFG motif and the B-site αD helix as conserved , coupled network nodes . The DFG motif is OUT when the KD is bound to the AD hub; an interaction analogous to the docking interaction of the PKA catalytic subunit with its regulatory subunit [7] . There were two distinct DFG IN orientations , one associated with activated and the other with inactivated states , as in Aurora kinase ( AurA ) [50] . The DFG motif is IN in auto-inhibited KD’s not bound to the hub . Thus , DFG motif orientation may report on compact and extended inactive holoenzyme states that have been visualized by EM cryo-tomography [51] . In contrast , helix αD is OUT in all auto-inhibited , but IN for activated open and substrate-bound structures . Our results add to data arguing that while the DFG motif is an important determinant of ATP binding its mobilization follows different strategies during activation of EPKs . In AurA , activation by the effector TpX2 involves an equilibrium population shift from the OUT to IN state [52]; but activation by phosphorylation triggers a switch between auto-inhibited and activated IN states [50] . Multiple strategies for kinase auto-inhibition have also been identified , for example , in the ZAP-70 tyrosine kinase [53] . Cooperativity in BCR-ABL , a tyrosine kinase identified as a hallmark for myeloid leukaemia can be either negative [54 , 55] or positive [26] depending on the coupling between the nucleotide and substrate binding sites . In CaMKII , as reported here , the nucleotide and substrate binding modules are again recruited for the formation of positive trans-subunit couplings . Comparative bioinformatics reveals common substrate binding site interactions between CaMKII and phosphorylase kinase [56] , but the phosphorylase holoenzyme is constitutively active and its complexity limits the study of nucleotide-substrate coupling . The histological , cellular and biochemical phenotypes of the human mutations that cause intellectual disabilities have been documented [18 , 19] . Brain imaging and electrophysiological recordings revealed cerebellar atrophy and abnormal action potentials ( APs ) respectively in many mutants . The common biochemical correlate was altered phosphorylation levels of the primary phosphorylation site ( T286 ) . T286 trans-phosphorylation has multiple short and long-term consequences for LTP . In the short-term second time scale , the optimal electrical stimulation frequency for LTP in neurons triggers dissociation from the actin cytoskeleton for sequestration at the PSD in dendritic spines . Synaptic localization of CaMKII holoenzymes by the triggered millisecond Ca2+ transients [57] persists over an hour [15] , implying the activated CaMKII transitions to a long-lived state . In vitro experiments report the Ca2+/CaM affinity for CaMKII R2 increases more than 1000-fold; essential for the frequency-dependent activation [29 , 58] . The activation triggers sub-second dissociation from actin and subunit exchange over minutes to propagate the activated state to previously inactive holoenzymes and prolong the lifetime of an activated holoenzyme population [59] . The phosphorylation also alters the affinity of some substrates for CaMKII , thus modifying substrate selection [60 , 61] . These multiple effects are all likely to arise from allosteric disruption of kinase cooperativity that would alter the optimal frequency [62] at which the holoenzyme responds to AP-triggered pulsatile calcium stimuli . Our MD simulations develop an initial allosteric framework responsible for the kinase cooperativity . Simulations of crystal structures of activated dimer complexes resolved changes triggered by dimerization and T286 phosphorylation . The match of the dimer RMSF to crystal B-factors indicated differential flexibility of the linked kinase domains . We found that conformational perturbation of the captured R1 helix drove the flexibility change to orchestrate the allosteric network . The R1 conformational perturbations reported by our simulations agree well with ESR analysis of R1 fragments in KDs [13] . The ESR study used an inactive , docked versus active , captured R1 two-state model to interpret the data . The free R1 would also form a third open state , as recognized in our comparative analysis ( Fig 6A ) , with similar ESR signature to the docked state; but the study did not examine the concentration dependence that could have distinguished between these states . The greater stability of the phosphorylated dimer is also consistent with ESR evidence that captured R1 lifetimes are increased in the phospho-mimic mutant T286E . T286 phosphorylation strengthened the coupling between chained KDs , with pK shifts of buried residues towards neutral values and surface , interfacial residues towards polar values , accompanied by compaction of the hydrophobic patch . The nanosecond coupling dynamics , relevant for the accommodation of the rapid changes in calmodulin and actin affinity , could be low-pass filtered in the dodecameric holoenzyme to freeze the activated cores over millisecond times after Ca2+/CaM dissociation to allow for diffusion of activated holoenzymes to the PSD . Similarly , during subunit exchange , the activated dimer must outlast the time for encounter with another holoenzyme . Diffusion and encounter times are on the order of milliseconds based on known spine volumes ( ~ 1 μm3 ) and spine CaMKII concentrations ( > 0 . 1 M ) . Conformational spread powered by strong inter-subunit coupling can achieve long-lived configurational states over seconds in multi-subunit ring assemblies such as the bacterial flagellar motor [63] . Structural elements that regulate the positive CaMKII kinase cooperativity have been identified . Reported Hill-coefficients ( H ) for substrate phosphorylation with Ca2+/CaM concentration varied from 4 . 3 to 1 . 1 in the C . elegans holoenzyme accompanied by substrate-dependent shifts in half-maximal dose . The H value is , for instance , downshifted by inhibitor peptides ( H 4 . 3 -> 1 . 7 ) . Importantly mutations ( F98E , I101D , I205K ) that introduce charged residues within or adjacent to the hydrophobic patch reduce dimer cooperativity in native / decoy KD mixtures ( H 1 . 5 -> 0 . 9 ) ; while the I -> D mutation in R1 residue 281D reduced holoenzyme cooperativity ( H 3 . 0 -> 1 . 5 ) [8] . Thus , available kinase cooperativity data agree qualitatively with the main features of the allosteric network we have characterized . A screen of tCONCOORD ensembles revealed the changes in the dimer allosteric network caused by in silico mutations TPO286A and I101D-I281D are similar and distinct from TPO286D ( S5 Fig ) . The change in kinase cooperativity with T286 phosphorylation , or H values for the phosphorylation ( TPO286D ) and dephosphorylation ( TPO286A ) mutant mimics , have not been reported . A triple alanine mutant ( T286-305-306A ) does not seem to affect cooperativity [8] but interpretation requires further work since phosphorylation of the inhibitory T305-T306 sites is antagonistic to T286 phosphorylation [1] . The relationship between inter-subunit coupling and cooperativity has been extensively modelled and experimentally characterized in the bacterial flagellar motor , a single ring assembly . Similar elucidation for CaMKII will be more challenging for this two ring dodecamer . Its architecture raises multiple possibilities for trans-phosphorylation [62] . Two distinct modes for conformational coupling have been proposed; a lateral spread of the activated conformation across the holoenzyme subunits [8] or transverse paired dimers [9 , 32] ( S6 Fig ) . The latter may also mediate activation-triggered subunit exchange via the central AD hub [64] . Interestingly . consistent with this idea , kinase cooperativity varies inversely with inter-domain linker length ( H 4 . 3 –> 1 . 7 ) [8] in the C . elegans holoenzyme and is reduced ( H 2 . 1 -> 1 . 1 ) by a mutation in the AD hub interface docking the KD DFG motif in the human holoenzyme [7] . The present study is an important first step towards this elucidation . It provides the first detailed framework of the allosteric network to reveal self-reinforcing R1 signal relays facilitate inter-subunit couplings that underlie any cooperativity mechanism . The mutated residue positions responsible for impaired human behaviour are distributed along the allosteric signal relays generated by R1 capture and strengthened by T286 phosphorylation . Thus , the probable molecular rationale for these mutations is that they principally act to disrupt the allosteric network rather than weaken substrate or nucleotide binding per se . Disruption of the allosteric network could have multiple outcomes as enumerated above to produce gain or loss of function with diverse pathophysiological consequences . Importantly , this study opens possible avenues for therapeutic treatment . MD simulations have proved useful for determination of the efficacy of peptide inhibitors to ATP-binding pocket residue mutations [65] . The design of allosteric inhibitors is more challenging as it requires an evaluation of the conformational plasticity of the protein assembly but promises greater specificity . We have identified protein sectors based on community analysis that may disrupt conformational coupling without altering calmodulin , nucleotide or substrate binding . Their predicted role as signal relays rather than binding determinants can be tested by mutant screens . An excellent example in the literature of such a mutation is PKA Y204A that disrupts coupling between nucleotide and substrate binding without affecting binding affinities [66] . In conclusion , our simulations make the case that the conformational dynamics of chained dimers have advantageous properties for subunit exchange and holoenzyme activation . Network analysis revealed the centrality of the coupled A ( DFG motif ) and B ( helix αD ) sites in the R1 relay to suggest how substrate affinity is modulated by nucleotide occupancy and how both influence cooperativity . It detailed how T286 phosphorylation strengthened conformational coupling initiated by R1 capture . Finally , community graphs identified targets for the rational design of allosteric inhibitors [22] . Future work will build on these advances to reconcile the dynamic architecture of the kinase with measurements of its cooperativity .
CaMKII sequences were retrieved from Uniprot [67] . MUSCLE was used for multiple sequence alignment ( MSA ) . Secondary structure predictions were made with PsiPred . The MSA was manually curated in Jalview . The phylogenetic tree was constructed with Fast-Tree 2 . 19 and displayed with Fig-Tree 4 . 3 ( http://tree . bio . ed . ac . uk/software/figtree/ ) . The human δKD structure alone ( PDB ID: 2VN9 ) and with calmodulin ( 2WEL ) [9] , complete subunit from the human holoenzyme ( 3SOA [7] ) ; the C . elegans KD structures alone ( 2BDW [23] , 3KK8 [8] ) and with CaMKIINtide ( 3KL8 [8] ) and the open form of protein kinase A ( 1CMK [68] ) were downloaded from Protein Data Bank . Missing atoms were added in Swiss-PDB viewer ( www . expasy . org/spdbv ) ; missing loop segments with Modeller ( https:/salilab . org/modeller ) . Mutant substitutions were made in Pymol ( http://pymol . org ) then energy minimized in Modeller . Parameters for tCONCOORD runs were as detailed earlier [69] . tCONCOORD utilizes a set of distance constraints , based on the statistics of residue interactions in a crystal structure library [25 , 70] , to generate conformational ensembles from an initial structure without the inclusion of solvent . Sets of 164 = 65 , 536 equilibrium conformations with full atom detail were typically generated for each structure . The overlap between ensemble subsets was > 99% when the subset size was < 1/4 of this value [69] . A set of 3 replicas of 300ns were generated for E . coli 3KK8 structure and its 286T equivalent using GROMACS 2016 . 2 with Amber ff99sb*-ILDNP force-field [71] . Each system was first solvated in an octahedral box with TIP3P water molecules with a minimal distance between protein and box boundaries of 12 Å . The box was then neutralized with Na+ ions . Solvation and ion addition were performed with the GROMACS preparation tools . A multistage equilibration protocol , modified from [72] as described in [33] , was applied to all simulations to remove unfavourable contacts and provide a reliable starting point for the production runs including: steepest descent and conjugate gradient energy minimisation with positional restraints ( 2000 kJ mol-1 nM-2 ) on protein atoms followed by a series of NVT MD simulations to progressively heat up the system to 300 K and remove the positional restraints with a finally NPT simulation for 250 ps with restraints lowered to 250 kJ mol-1 nM-2 . All the restraints were removed for the production runs at 300 K . In the NVT simulations temperature was controlled by the Berendsen thermostat with coupling constant of 0 . 2 ps , while in the NPT simulations the V-rescale thermostat [73] was used with coupling constant of 0 . 1 ps and pressure was set to 1 bar with the Parrinello-Rahman barostat and coupling constant of 2 ps [74] . A set of 3 replicas with time step 2 . 0 fs with constraints on all the bonds were used . The particle mesh Ewald method was used to treat the long-range electrostatic interactions with the cut-off distances set at 12 Å . The threonine phosphate ( TPO286 ) was changed to threonine ( T286 ) after equilibration to generate the non-phosphorylated form . The 300 ns MD runs reached stationary root mean square deviation ( RMSD ) values within 3 ns . Contact residues , surfaces and energies were extracted from the PDB files with the sub-routines ( ncont , pisa ) available in CCP4 version 7 ( http://www . ccp4 . ac . uk/ ) . Continuum electrostatics were computed with the Poisson Boltzmann solver ( http://www . poissonboltzmann . org/ ) [75] . Comparison with experimental B-factors and geometrical analyses were performed with GROMACS version 4 . 5 . 7 ( http://www . gromacs . org/ ) . The mutual information I ( X;Y ) between two variables ( X ) and ( Y ) is I ( X;Y ) =H ( X ) +H ( Y ) −H ( X , Y ) ; where H ( X , Y ) is the joint probability distribution; The normalized mutual information , nMI ( X;Y ) = ( I ( X;Y ) −ε ( X;Y ) ) / ( H ( X , Y ) ) ; H ( X ) is a measure of the entropy ΔS ( X ) that is related to the number of microstates and their probability . kB Is the Boltzmann constant ΔS ( X ) =kB . ln ( WX ) =kB . ∑i=1np ( Xi ) . ln ( p ( Xi ) ) H ( X ) =ΔS ( X ) /kB . ε ( X;Y ) is the expected , finite-size error . The finite-size error estimated as in earlier publications ( e . g . [45 , 76] ) corrects for the effects of finite data and quantization on the probability distribution [77] . The nMI couplings are detected as correlated changes in fragment dynamics , after spatial filtration to isolate long-range couplings [76] . In methods based on the dynamics cross-correlation matrix [78] , the correlation is calculated from position fluctuations of the residue Cα atoms . In the MutInf method [79] , correlations are measured in terms of mutual Information between discretized torsional angles . The residue-based approaches do not directly read-out couplings between structural motifs , e . g . secondary structures . The SA [80] , is a set of recurring four residue fragments encoding structural motifs derived from PDB structures . There is no need for discretisation and / or optimisation of parameters as the fragment set is pre-calculated . Collective motions were identified by PCA of the conformational ensembles . PCs were generated by diagonalization of the covariance matrix of Cα positions in GROMACS 4 . 5 . 7 . The overlap ( cumulative root mean square inner product ) of the PCs between replicas [40] ) and the PC dot product matrix was computed with the GROMACS g-anaeig function . The motions have no time-scale for tCONCOORD ensembles , but comparison with MD trajectories was consistent with the notion that collective motions represented by the first few PCs are “slow” relative to smaller amplitude motions recorded by the later PCs . Statistically significant correlations between columns were identified with GSATools [36] and recorded as a correlation matrix . The correlation matrix was used to generate a network model; with the residues as nodes and the correlations as edges . The contribution of a node to the network was estimated by the eigenvector centrality , E , calculated directly from the correlation matrix: E . [M]corr=E . λ where [M]corr is the correlation matrix and λ is the eigenvalue The Girvan–Newman algorithm [37] was used to identify community structure . Then the network was collapsed into a simplified graph with one node per community , where the node size is proportional to the number of residues . Edge weights represent the number of nMI couplings between communities [38] . Community analysis of correlation networks identifies relatively independent communities that behave as semi-rigid bodies . Graphs were constructed with the igraph library [81] in R ( https://cran . r-project . org/web/packages/igraph/ ) and visualized in Cytoscape ( http://www . cytoscape . org/ ) . The typical size of a tCONCOORD ensemble was 65 , 536 conformations ( 2562 ) . Three MD replicate runs for the two ( phosphorylated , dephosphorylated ) conformations . Ensemble conformations and MD runs were averaged for computation of the nMI between fragment positions , with > 2σ threshold for selected top couplings . Pearson’s correlations were used for comparison . Significance limits were set in GSATools . S1 Table lists the web databases and software servers used in this study . | Protein kinases play central roles in intracellular signalling . Auto-phosphorylation by bound nucleotide typically precedes phosphate transfer to multiple substrates . Protein conformational changes are central to kinase function , altering binding affinities to change cellular location and shunt from one signal pathway to another . In the brain , the multi-subunit kinase , CaMKII is activated by calcium-calmodulin upon calcium jumps produced by synaptic stimulation . Auto-transphosphorylation of a regulatory peptide enables the kinase to remain activated and mediate long-term behavioural effects after return to basal calcium levels . A database of mutated residues responsible for these effects is difficult to reconcile solely with impaired nucleotide or substrate binding . Therefore , we have computationally generated interaction networks to map the conformational plasticity of the kinase domains where most mutations localize . The network generated from the atomic structure of a phosphorylated dimer resolves protein sectors based on their collective motions . The sectors link nucleotide and substrate binding sites in self-reinforcing relays between regulatory peptides . The self-reinforcement is strengthened by phosphorylation consistent with the reported positive cooperativity of kinase activity with calcium-calmodulin concentration . The network gives a better match with the mutations and , in addition , reveals target sites for drug development . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] | [
"phosphorylation",
"crystal",
"structure",
"monomers",
"condensed",
"matter",
"physics",
"materials",
"science",
"network",
"analysis",
"sequence",
"motif",
"analysis",
"crystallography",
"oligomers",
"research",
"and",
"analysis",
"methods",
"sequence",
"analysis",
"polymer",
"chemistry",
"computer",
"and",
"information",
"sciences",
"solid",
"state",
"physics",
"bioinformatics",
"proteins",
"chemistry",
"centrality",
"physics",
"biochemistry",
"biochemical",
"simulations",
"post-translational",
"modification",
"database",
"and",
"informatics",
"methods",
"biology",
"and",
"life",
"sciences",
"physical",
"sciences",
"dimers",
"materials",
"computational",
"biology"
] | 2019 | Conformational coupling by trans-phosphorylation in calcium calmodulin dependent kinase II |
Melioidosis is an often fatal infectious disease affecting humans and animals in tropical regions and is caused by the saprophytic environmental bacterium Burkholderia pseudomallei . Domestic gardens are not only a common source of exposure to soil and thus to B . pseudomallei , but they also have been found to contain more B . pseudomallei than other environments . In this study we addressed whether anthropogenic manipulations common to gardens such as irrigation or fertilizers change the occurrence of B . pseudomallei . We conducted a soil microcosm experiment with a range of fertilizers and soil types as well as a longitudinal interventional study over three years on an experimental fertilized field site in an area naturally positive for B . pseudomallei . Irrigation was the only consistent treatment to increase B . pseudomallei occurrence over time . The effects of fertilizers upon these bacteria depended on soil texture , physicochemical soil properties and biotic factors . Nitrates and urea increased B . pseudomallei load in sand while phosphates had a positive effect in clay . The high buffering and cation exchange capacities of organic material found in a commercial potting mix led to a marked increase in soil salinity with no survival of B . pseudomallei after four weeks in the potting mix sampled . Imported grasses were also associated with B . pseudomallei occurrence in a multivariate model . With increasing population density in endemic areas these findings inform the identification of areas in the anthropogenic environment with increased risk of exposure to B . pseudomallei .
Southeast Asia and tropical Australia have recently experienced a surge in melioidosis , an often fatal infectious disease caused by the saprophytic environmental bacterium Burkholderia pseudomallei [1 , 2] . Case numbers in the Top End of Australia have substantially increased in recent years . In the 20 years from 1989 until 2009 there was a median of 27 cases annually [3] . In the last 5 years there has been a median of 64 cases annually and in each of two recent years , 1 in every 2 , 000 people living in the Top End has had culture confirmed melioidosis [4] . B . pseudomallei are found in soil and water world-wide in the tropical belt with the major endemic region being southeast Asia and tropical Australia [5–10] . B . pseudomallei is an opportunistic pathogen able to infect humans [11] and a large variety of animals [12] . Humans with a compromised immune system such as from diabetes , hazardous alcohol use , chronic renal disease and immunosuppressive therapy are at particular risk of acquiring and dying from melioidosis [13] . Clinical presentations vary widely and include skin and soft tissue abscesses , pneumonia and disseminated infection with septic shock , the latter having mortality rates above 80% [14] . The Darwin area ( 12° S latitude ) in the tropical north of Australia is endemic for melioidosis and gardening is considered to be an important recreational and occupational source of exposure to and ultimately , infection with B . pseudomallei [3] . In the 20-year Darwin prospective melioidosis study , 407 ( 75% ) of 540 consecutive melioidosis patients had documented recreational activities such as gardening or outdoor sporting activities where exposure to B . pseudomallei was considered likely to occur [3] . Domestic gardens are not only a common ground for humans to be exposed to the environment , but B . pseudomallei might also thrive in the garden habitat . While B . pseudomallei and melioidosis predominate in the monsoonal wet season [3] , previous work in rural Darwin found that in the dry season B . pseudomallei is more often present in domestic gardens than in farms or environmentally less disturbed areas [15] . This might be attributed to the widespread use of irrigation during the dry season . Being a non-spore forming , gram negative bacterium , B . pseudomallei is often , but not exclusively associated with moist soil close to a water source and with surface water or alluvial areas as well as rice fields [7 , 15–19] . At environmentally disturbed sites , B . pseudomallei was associated with pens or paddocks for pigs , chickens or horses with an average odds ratio of 3 . 8 [15] . This raises the possibility that soil aeration through digging activities or organic material and nitrogen from animal waste support growth of B . pseudomallei [15] . In this study , we addressed the hypothesis that anthropogenic manipulations associated with gardens such as the use of irrigation , fertilizers , commercial potting mix or keeping pets influence the habitat of B . pseudomallei and change its abundance and/or occurrence . We conducted a soil microcosm experiment with a selection of fertilizers as well as a longitudinal study over three years on an experimental fertilized field site in a location naturally endemic for B . pseudomallei .
In August 2008 an experimental site was established on a private property in rural Darwin in an area that previously tested positive for B . pseudomallei . The soil at this site was a hydrosol [20] and the soil texture of the topsoil was clay with a subsoil consisting of grey clays and siltstone . The site consisted of two plots , 0 . 75 metres apart and each plot had six 1x1 metre quadrants ( Fig . 1 ) , which included a control quadrant and five quadrants with different treatments which represent common garden practices in the Darwin region ( Table 1 ) . Treatments were applied every two weeks with water application every 2nd day for three years . Timing and dose reflected local garden practices . The water used was unchlorinated water from the property’s bore with a pH of 7 . 5 containing 50 mg/L calcium carbonate and which repeatedly tested negative for B . pseudomallei by culture . There were 14 rounds of soil sampling and in each round 2 random soil samples were collected from each of the 12 quadrants to give a total of 336 soil samples . The first sampling round was before the start of the experiment in August 2008 followed by sampling every two months in year-1 , every three months in year-2 and every four months in year-3 of the experiment , with the last round in August 2011 . Soil from a depth of 20–30 cm was collected into sterile 50 mL specimen containers and auger and spade were cleaned with 70% ethanol between soil collections [21] . Soil moisture was determined as described previously using the Australian Soil and Land Survey Field Handbook [21 , 22] . Soil pH was measured using a soil pH field test kit ( Inoculo , Australia ) . In the last 6 months , soil electrical conductivity ( EC ) was measured using the Field Scout EC Meter ( Spectrum Technologies , USA ) . Grasses covering the experimental field site were identified by the Northern Territory Government Herbarium and were either Sorghum spp . ( spear grass ) or Pennisetum pedicellatum ( annual “mission grass” ) . At the time of sampling , the presence or absence of live specimens of these grasses at the sampling hole was noted . Of 120 250-mL clean and autoclaved plastic containers , 30 were each filled with either 130 g of commercial “garden soil” , sandy clay loam , clay or sand ( Table 2 ) . The non-commercial soil was collected in rural Darwin and tested negative for B . pseudomallei by culture . Nine different treatments plus controls ( no change ) were applied in triplicate to the containers ( Table 3 ) . Treatments included the addition of distilled water or distilled water in combination with eight fertilizers which are commonly used in residential gardens in the Darwin region . After two weeks of soil conditioning at 32 degrees Celsius in the dark , all soils were inoculated with 5x10e4 CFU of an environmental strain of B . pseudomallei ( MSHR2817 ) which has the commonly found multi-locus sequence type ( ST ) 144 [23] and incubated at 32 degrees Celsius for four weeks in the dark . Soil DNA was extracted and B . pseudomallei DNA detected as described below . Soil DNA extraction was done as previously described [15 , 21] . Briefly , 20 g of soil were incubated with 20 mL of Ashdown’s broth for 39 hours shaking at 37°C , the soil supernatant was centrifuged twice and the pellet processed using the PowerSoil Kit ( MoBio Laboratories , USA ) . Modifications included the addition of 0 . 8 mg of aurintricarboxylic acid ( ATA ) and 20 μL of proteinase K ( 20 mg / mL ) . B . pseudomallei DNA was targeted by the well validated B . pseudomallei specific Type Three Secretion System-1 TTS1 real-time PCR [24 , 25] . For the microcosm experiment , DNA was extracted from 20 g of soil using the previously described semi-quantification method with an internal extraction and amplification plasmid control [23] . TTS1 copy numbers were normalized by dividing them by the copy number of the internal pt7 plasmid control which was added to the soil samples prior to extraction , in order to account for differences in DNA extraction and amplification efficiency as a result of varying amounts of inhibitors present in soil samples [23] . Statistical analysis was carried out using Stata ( Intercooled Stata , version 12 . 1 , USA ) . For bivariate analyses , Fisher’s exact test and Mann-Whitney U test were used . All tests were 2-tailed and considered significant if P values were less than 0 . 05 . Graphs were generated in Stata and GraphPad Prism 6 . For the experimental field site , a conditional logistic regression model was used to model the odds of B . pseudomallei occurrence once the experiment had started , with fixed effects for treatments and dates of sampling . Fractional treatment effects were assumed for the first 12 months ( e . g . 50% of full treatment effect after 6 months ) to allow the application of fertilizers to have a gradual effect on the soil environment and Burkholderia community [26] . Heat maps for the experimental field site were generated using the thin-plate-spline interpolation method in ArcGIS 10 . 1 ( ESRI 2012 ) .
B . pseudomallei occurrence was monitored over three years on a field site with five different treatments applied in an area in rural Darwin naturally positive for B . pseudomallei ( Fig . 1 ) The microcosm experiment was used to determine whether commercial fertilizers commonly used in gardens in the Darwin region increased B . pseudomallei load in soil . Eight different treatments of garden fertilizers were applied to each of four different soil types in triplicate . There were also triplicate water controls with only distilled water added and triplicate controls with nothing added . Four weeks after inoculation , no B . pseudomallei were detected in commercial garden soil for any of the treatments . Sand and clay contained on average 733 times more B . pseudomallei than sandy clay loam which only contained minimal B . pseudomallei cells ( Fig . 5 ) . After four weeks , no B . pseudomallei ( 10/12 ) or only minimal B . pseudomallei cells ( 2/12 ) were detected in the 12 control samples . The effect of fertilizers upon B . pseudomallei differed between soil type ( Fig . 5 ) . In sand , the addition of a fertilizer rich in urea showed the highest B . pseudomallei load compared to controls but the same fertilizer only had a moderate effect in clay and no effect in sandy clay loam . The pH and EC of urea in sand were lower with a mean of 4 . 6 and 56 μS/cm in comparison to sandy clay loam ( pH 5 . 8 and EC 216 μS/cm ) and clay ( 8 . 4 and 172 μS/cm ) . The addition of an organic and a phosphorus rich fertilizer resulted in the two highest mean B . pseudomallei counts in clay but no such effect was seen for the other soils . The addition of water alone caused a similarly high increase in B . pseudomallei load in sand and clay which had a low VSW of 0–2% before addition of water but there was no load increase in sandy clay loam with a higher initial VSW of 5% .
In summary there was clear evidence for irrigation increasing B . pseudomallei occurrence . The effect of fertilizer application upon B . pseudomallei was more complex and was dependant on soil type and physicochemical properties as well as on vegetation , with nutrients also causing an increase in plant root development beneficial to B . pseudomallei . The use of fertilizers is causing drastic changes to the global nutrient cycle with a significant rise in supply of otherwise limiting nutrients . These changes have a major impact upon the soil and water microbial community structure and likely also upon host pathogen interactions [50] , including those involving B . pseudomallei . | Melioidosis cases are on the rise in endemic areas of northern Australia and Thailand . This potentially severe infectious disease affecting humans and animals in the tropical belt is caused by the gram negative bacterium Burkholderia pseudomallei . Domestic gardens are a common point of exposure to these environmental bacteria and B . pseudomallei are more prevalent in the dry season in gardens when compared to other areas . This is why we analysed whether common gardening practices such as regular watering ( irrigation ) or soil fertilizing change the occurrence of B . pseudomallei . We conducted a soil microcosm experiment with a range of fertilizers and soil types as well as a longitudinal interventional study over three years on an experimental fertilized field site in an area naturally positive for B . pseudomallei . Irrigation was the only consistent treatment to increase B . pseudomallei occurrence over time . The effects of fertilizers upon these bacteria depended on soil texture , physicochemical properties such as pH or salinity and vegetation . B . pseudomallei occurrence was also associated with imported grasses . With increasing populations in endemic areas , these findings inform the identification of areas in the anthropogenic environment with increased risk of exposure to B . pseudomallei . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] | [] | 2015 | What Drives the Occurrence of the Melioidosis Bacterium Burkholderia pseudomallei in Domestic Gardens? |
The Pak-Mun dam is a controversial hydro-power project on the Mun River in Northeast Thailand . The dam is sited in a habitat of the freshwater snail Neotricula aperta , which is the intermediate host for the parasitic blood-fluke Schistosoma mekongi causing Mekong schistosomiasis in humans in Cambodia and Laos . Few data are available which can be used to assess the effects of water resource development on N . aperta . The aim of this study was to obtain data and to analyze the possible impact of the dam on N . aperta population growth . Estimated population densities were recorded for an N . aperta population in the Mun River 27 km upstream of Pak-Mun , from 1990 to 2011 . The Pak-Mul dam began to operate in 1994 . Population growth was modeled using a linear mixed model expression of a modified Gompertz stochastic state-space exponential growth model . The N . aperta population was found to be quite stable , with the estimated growth parameter not significantly different from zero . Nevertheless , some marked changes in snail population density were observed which were coincident with changes in dam operation policy . The study found that there has been no marked increase in N . aperta population growth following operation of the Pak-Mun dam . The analysis did indicate a large and statistically significant increase in population density immediately after the dam came into operation; however , this increase was not persistent . The study has provided the first vital baseline data on N . aperta population behavior near to the Pak-Mun dam and suggests that the operation policy of the dam may have an impact on snail population density . Nevertheless , additional studies are required for other N . aperta populations in the Mun River and for an extended time series , to confirm or refine the findings of this work .
The Pak-Mun dam in Northeast Thailand , has been the focus of much controversy since the announcement of plans for its construction in 1982 . Demonstrations and protests against the dam became more pronounced in 1989 when the project received government approval [1] . Construction of the dam began in June 1990 and the project was completed on 26th June 1994 . Commercial operation of the 136 MW hydroelectric plant began in November 1994 , but by the 9th October 1994 all four of the dam's generators were already on line and under test . The Pak-Mun project has been heavily criticized over the associated risk or flooding , displacement of villages and damage to tourist areas such as the rapids at Kaeng-Sa-Poe [1] , but it is the risk of schistosomiasis transmission which is the concern of the present study . The Pak-Mun Dam is located at Ban Hua Heo in Ubon Ratchathanee Province on the Mun river ( 15 . 281981 N 105 . 468095 E ) , 5 . 5 km upstream of its confluence with the Mekong river at the Lao:Thai border ( Figure 1 ) . The rocky areas around small islands in this section of the Mun river are the habitats of the epilithic rissooidean freshwater snail Neotricula aperta ( Pomatiopsidae: Triculinae ) . Neotricula aperta is the intermediate host for the helminth blood-fluke parasite Schistosoma mekongi ( Trematoda: Digenea ) in the Mekong river and associated tributaries of Laos and Cambodia . Three strains of N . aperta are recognized ( α , β and γ , respectively , on the basis of decreasing shell length ) , all three strains are able to transmit the parasite in the laboratory , but in nature only the γ-strain is found to be epidemiologically significant . The β-strain is known only from the lower Mun river and is the only strain to be found in this river ( Figure S1 ) . The average shell length of the β-strain is less than 3 mm . In laboratory studies N . β-aperta has been found to be the most susceptible strain to infection with S . mekongi [2] , although this may vary depending on the source of parasite used [3] . Schistosoma mekongi infects humans in Laos and Cambodia and causes Mekong schistosomiasis ( at foci located about 190 km from the Pak-Mun Dam , see Figure 1 ) . Currently , N . aperta is known from 31 localities in Cambodia , Lao PDR and Thailand , involving nine river systems [4] and an estimated 1 . 5 million people are at risk of infection by S . mekongi [5] . Neither the disease nor infected snails have been reported from the Mun river; however , the possibility that the dam might alter ecological conditions so as to favor transmission of the parasite has been one of the main concerns regarding the Pak-Mun project [1] , [6]–[9] . The major rivers of the lower Mekong Basin show a marked high water and low water seasonal flow pattern , with N . aperta completing its life-cycle within the low water period ( March to May annually ) . Neotricula aperta is found only in shallow areas ( typically 0 . 5 to 3 m deep ) of these rivers . The snails are restricted to areas where the current is moderate ( around 2×103 m3 s−1 ) , the water is clear , well oxygenated , and the bed rock forms ( almost flat ) platforms where algal aufwuchs is extensive [10] . Such conditions exist only during the low water period; therefore N . aperta persists mostly by recruitment ( from eggs laid on stones in the previous year ) or re-colonization from other rivers [11] . Neotricula aperta grazes the algal epilithon and cannot survive in areas where sediment is depositing and preventing the growth of the algae upon which it feeds . Indeed , ecological studies of N . aperta have shown that this snail is restricted to stones covered with fine sediments and that this species is highly sensitive to silting [12] . These “prosobranch” snails ( Rissooidea ) show ecological requirements that are very different from those of the pulmonate snails which transmit schistosomiasis to humans in Africa and South America; these differences have been the basis of calls for specific models to predict the effects of dam projects on the transmission of S . mekongi , because lessons learned from Africa and South America may not be valid in the case of Mekong schistosomiasis [13] , [14] . The Pak-Mun dam is comprised of roller compacted concrete with a maximum height of 17 m and total length of 300 m . The reservoir has a surface area of 60 km2 , at the average high water level of 108 m above the mean sea level ( MSL ) , and a capacity of 225 million m3 . The operating rules of the dam are designed to ensure that the water level does not exceed 106 m MSL during the low water season [1] . Consequently , the ecological conditions of the river are not much affected during the high water season but they are more greatly impacted during the low water period , which is the time when N . aperta is active . Following intense protest by riparian villagers in 2001 , the Thai government agreed to open the gates of the Pak-Mun Dam for one year to conduct studies on the impacts of the dam on fisheries , social life and electricity supply . The study was conducted between June 2001 and July 2002 and the dam was re-closed from 1 November 2002 onwards , although the dam gates were allowed to be left open from July to October each year , thereafter , to cover the height of the flood season [15] . In spite of the many articles on line and in the popular press , which discuss the problems associated with the Pak-Mun dam project , there are very few peer-reviewed formal scientific studies available . Indeed there are only two studies on snail population density near Pak-Mun from 1996 [8] , [9] and one study screening local people for Mekong schistosomiasis in 2004 [7] . The impact of the dam clearly includes attenuation of the effects of the annual flood/dry season cycle , with higher water levels , lower flow rates and ( consequently ) higher rates of silt deposition in the low water period . Such changes might be expected to lower N . aperta population densities; however , a reduction in flow rate between May and July might allow extension of the snails' breeding period and lead to an increase in population density . It is important that we understand the likely impacts of water resource development on N . aperta populations because plans are under consideration for at least 12 hydroelectric power dams on major rivers in the lower Mekong Basin and much controversy surrounds predictions of their environmental impacts [16]–[18] . Fortunately , one of the 1996 studies on N . aperta near Pak Mun [9] was performed by the present authors and used consistent snail density estimation techniques , allowing the data to be combined with those of the present study to form a time series , with frequent samples , running from 1990 to 2011 . All samples were taken at a series of small islands in the Mun river near Ban Hin Laht ( 15 . 227774 N 105 . 271035 E ) located 27 km upstream of the dam ( Figure S2 ) . The β-strain of N . aperta is no longer found downstream of the dam , although the γ-strain and many Manningiella polita snails ( M . polita is a close relative of N . aperta and could be confused with N . β-aperta ) are found at the confluence of the Mekong and Mun rivers . During the late 1990s it was not possible to take samples close to the Pak Mun dam due to political unrest . The 1996 study covered 1991 to 1995 and used ANOVA to detect any density differences among sample years , followed by pairwise comparisons between years using simple non-parametric tests . At that time no suitable method was available for analysis of short time series . Nevertheless , the study did find a statistically significant increase in densities observed after the dam came into operation compared with those seen before operation of the dam . Unfortunately , the study terminated only one year after the dam was completed and so it could give no indication of the long-term impact of the dam on snail population density . Trends can be estimated for an observed time series using a linear mixed model expression of the Gompertz stochastic state-space exponential growth model . Modifications of this model , in the form of a linear , normal state-space model , are now available to allow for unequally spaced sampling intervals [19] , [20] . The resulting model thus provides a method of trend analysis incorporating both observation error and environmental process noise , which has proven suitable for use with shorter time series and uneven sampling times ( as encountered in the present investigation ) [19] . Popular approaches such as ARIMA are not appropriate in the present case because these require 50 or more regular samples [21] . The approach has already proven useful in studies of N . aperta population changes around the Nam Theun 2 dam site in Laos [14] . In view of the unique biology and ecological niche of N . aperta ( relative to pulmonate snails in Africa or South America ) , the lack of data upon which to base models , and the uncertainty as to the likely response of the snail populations to the Pak-Mun dam , further studies are clearly necessary . Consequently , the present work was undertaken in order to help fill this gap in our understanding and to investigate N . aperta population density changes ( if any ) occurring at Ban Hin Laht in the 19 years following the first operation of the Pak Mun dam . In this study snail population density estimates were collected and subjected to a trend analysis using the modified Gompertz State-Space exponential growth model . The objective was to detect any population trend ( either growth or decline ) and to look for any significant changes concurrent with operation of the dam .
Snails were collected off stones submerged in the Mun river at Ban Hin Laht ( 15 . 227774 N 105 . 271035 E ) , the collections were a continuation of those begun in 1996 ( refer to site 1 , note the 1996 GPS coordinates were inaccurate ) [9] . Sampling was annual or biennial for 1990–1995 and 2000–2005 , with a final sample in 2011 . All samples were taken within the low water period of the Mun river and within a five week period of each sampled year , before the onset of the spate which would affect the distribution of snails in the river and submerge the islands around which the populations were sampled . The sample site was chosen because it was an area , out of the sites originally sampled in the 1996 study , of relatively low current , shallow waters and did not infringe on the fishing activities of other villagers or the boat passage ( established since 1996 ) . Consequently , the site was less risk to collector safety but covered 5480 m2 of river bed ( measured by GPS ) . The area was approximately square , with five samples taken from within , one at each corner and one in the centre . The samples were taken at depths within the range 0 . 8 to 1 . 5 m . The sub-sample points were originally marked by lining up landmarks and cross referencing , but after 1999 by GPS . The beta strain of N . aperta is almost exclusively epilithic and is never found on or in mud or on aquatic macrophytes , consequently the snail populations could be sampled off stones collected from the river . A diver entered the water at each point and collected stones in the 113–1922 cm2 ( surface area ) range , such that approximately 0 . 5 m2 was sampled at each point . The stones were lifted out of the water directly into a plastic tray and carried back to the boat . The stones , in the trays , were left exposed to the sun for five minutes and then the stones were turned over and exposed for a further 5 minutes . The stones were then doused in water; this has the effect of causing the N . aperta to drop from the stones ( they will otherwise try to lock the aperture of their shell to the stone and be difficult to find and dislodge ) . The stones were then removed , re-exposed to the sun and washed in new trays to ensure that all snails had been removed . The trays were then left in the shade for 3 minutes , during which time the snails attached themselves to the tray . The trays were then washed of any mud , sand or insects , filled with clean river water and the N . aperta present removed and counted . The surface area of the stones was later measured by covering the stone with plastic food wrap ( cutting it to fit in a tight mono-layer ) ; the plastic wraps were labeled and later weighed in the laboratory and a weight calibration factor used to determine the surface area . The area of any stones collected which bore no N . aperta was also taken into account . To further standardize the procedure all samples were taken between 10:15 am and 11:30 am . The same field personnel* were used throughout the study and the diver was restricted to only 30 seconds searching time at each point ( to deter from the selection of stones with most snails attached or large stones ) . *In 2003 the diver was replaced by his son so as to retain the ability to sample the heavier stones . The sampling approach involved two main assumptions; first that sampling off the stones is representative of the population density elsewhere on the river bed , and second that the site chosen is representative of all other sites along the river near to the dam . Considering the first assumption , the river bed at Ban Hin Laht is comprised of large blocks and ridges of rock incised by deep crevices . Snails are likely to cover the sides of these crevices down to the depths of sufficient light penetration , and so the true area of the sampling site is vastly greater than 5480 m2 . In view of this , sampling off collected stones is far more representative than counting snails per unit area of sample site . It is likely that the density on the stones collected is proportional to the density on the bed rock upon which they rest because there is evidence that any stone placed in the river is rapidly colonized by N . aperta [22] , thus snails appear to move freely between stones and other substrata . With regard to the second main assumption , the 1996 study examined inter-site variation in snail density among five sites around Ban Hin Laht ( including the site of the present study ) and found no significant variation ( Kruskal-Wallis one-way ANOVA ) . All analyses in this study were performed using the R statistical package [23] . The sample dates were converted to weeks , with the first sample being designated as week zero . In order to determine if the counts showed a simple linear relationship with time and to assess the effects of any outliers , a simple linear regression was performed for the data . The data ( snail counts per m2 ) were then screened for outliers using plots of contour lines for the Cook's distance ( a measure of the influence of each data point on the overall regression result ) , with distances >4/n ( where n is the number of observations ) suggesting the presence of a possible outlier [24] . In order to check whether the data might be normally distributed a Shapiro-Wilk test was applied . Similarly , the fit of the data to a Poisson General Linear Model ( GLM ) and to a negative binomial GLM were also assessed . The methods used follow those of an earlier study [14] , but there follows a brief outline of the approach and details of starting parameters , replicates and other conditions specific to the present study . Initially the observed N . aperta population densities were fitted to a modified Gompertz State-Space exponential growth model ( GSS ) [20] , [25]–[27] and relaxed maximum likelihood estimates ( REML ) obtained for the four unknown parameters of the model under the standard model of deterministic exponential growth; these being:where β is the hypothesized ( unobserved ) initial value of the ( log ) population density ( at time t = 0 ) , μ is the expected change in population density over one ( sampling ) unit of time , σ2 is a variance parameter representing process noise ( environmental variability ) , and τ2 is a variance parameter accounting for observation error . The primary aim is to estimate μ ( or mu ) , the trend parameter , which is the rate at which median population density changes over time . Two of the most commonly applied methods in time series analysis are the LLR model , log-linear regression of counts against time , and the Stochastic Model of Exponential Growth ( SEG ) . The GSS is preferable to both these models . The SEG incorporates process noise and yields a log-normal probability distribution of population density , but it assumes that variability in abundances is due entirely to growth rate fluctuations caused by environmental variability ( process noise ) . The LLR model implicitly assumes that observation error is the sole source of variation in the data ( with population density governed by deterministic exponential growth ) . In contrast , the GSS allows for both environmental variability and sampling error . Nevertheless , the LLR and SEG models were able to provide the starting values for the four parameters of the GSS at which the REML searches were initiated . The initial value of μ ( mu0 ) was the average of the estimate from the two models , the initial σ2 ( ssq0 ) was from the SEG model , and the initial τ2 ( tsq0 ) and X0 ( X00 ) from the LLR model . The fit of the scaled GSS model was maximized using the R script for the multivariate normal log-likelihood function provided in the literature [19] , which uses Nelder-Mead optimization . Ten runs of 100000 iterations each were executed with initial parameter values drawn from a normal distribution , with mean equal to the estimates from the LLR and SEG models and a SD of 1; the random number seed was also changed between each run . The REML value was noted from these runs and ten more runs performed , but scripted to only return values corresponding to a better REML . If no improvement in REML was observed , the Nelder-Mead searches were discontinued and the analysis proceeded to the McMc step described below . GSS models can lead to a likelihood surface with multiple local maxima , such that the chance of detecting the true peak using a non-global method such as Nelder-Mead is unlikely unless the initial parameter values are very close to this peak [28] . To overcome this problem a simulated-annealing approach linked to a Metropolis sampler ( SAN ) was used to better explore the likelihood surface . Published studies had reported some success using the SAN approach in this context [20] . The SAN searches were run using the implementation in R ( details elsewhere [14] ) for 100 million generations . The model fitting was also repeated 10 times , with initial parameter values drawn from a normal distribution , with mean equal to the starting values from the LLR and SEG models and SD set such that 95% of the values drawn lay within 0 . 1 or 10 times the mean . The random number seed was also varied between runs . A REML frequency histogram was then plotted and the parameter estimates for models , corresponding to the minimum , mean , modal and maximum REML values , were inspected for relative variation and biological credibility . In addition , profile likelihoods were plotted ( for a range of fixed mu values ) to ensure that the procedure was effecting a stable and reliable estimation of the trend parameter . The parameter estimations were repeated , using the previous “best” REML estimates as starting values , until ten runs could be performed without any improvement in likelihood or change in estimated parameter values . In order to determine if the snail populations showed any long-term trend ( either growth or decline ) a test for zero trend was performed using the standard error of the estimated mu and a standard normal percentile in an equivalence testing framework [29] . The null hypothesis being that a significant trend is present ( i . e . , the mu estimate lies outside a fixed , specified , interval containing zero ) . The equivalence region was obtained by simulating the data 1 million times , with mu set to zero and then allowing the GSS ( with SAN ) to provide REML estimates of mu for these simulated data sets . The simulation procedure was that used by previous authors [19] . The starting values for the other three parameters were the REML estimates ( see above ) , with the error terms ( E and F ) drawn from normal distributions with mean zero and SD ssq and tsq , respectively . The exponential growth equation is then ( on the log scale ) :where Yt represents our observations at time t , whilst Xt represents the true population density that we cannot observe ( only sample ) , and the simulated population densities are exp ( Yt ) . The 95% confidence interval ( CI , i . e . , ±t . SE ( mu ) ) for the 1 million mu values was then used as the equivalence region . A naïve bootstrap procedure was used to generate a 97% CI for mu ( equivalent to a 95% CI in a two-tailed test ) . The original data were bootstrapped 1 million times using the “sample ( ) ” function in R ( i . e . , simple sampling with replacement ) , the time series was re-ordered by week , and if any two weeks ( observations ) had the same value , one week was added to the latter observation . REML estimates were then obtained using the GSS and SAN as described above . If the 95% CI for the empirical ( bootstrapped ) mu lay entirely outside that for the simulated mu values , then the null hypothesis that a substantial trend is present would be tentatively accepted . The procedure is not ideal because the same model is used to estimate mu and to simulate the data for the equivalence region , also the time series is short and the empirical cumulative distribution function ( cdf ) after bootstrapping may not be a good approximation to the true cdf . Nevertheless , the method avoids the subjective decisions of conventional equivalence testing , such as what the upper and lower bounds of the equivalence region should be for a snail population little studied previously in this respect . It was considered necessary to derive a work flow by which population densities could be predicted and CIs obtained for these predictions . A Kalman filter was used to yield optimized estimates of population density according to the scaled GSS model and REML parameter estimates . The filter estimates the current value of the population density at time t ( Xt ) under the scaled GSS , given the history of the Yt values up to and including Yt [20] . The R script used to implement the filter was taken from the literature [19]; however , this procedure assumes that the population under study is far from equilibrium ( e . g . , it is relatively recently established ) . The history of the N . aperta populations in the Mun river is not known and the populations could have been fluctuating around the habitat's carrying capacity for some time . Consequently , the Kalman filter was modified to assume stationarity [20] and predictions for both stationary and non-stationary cases were made ( here stationary refers to the joint distribution of all Yt ) . The CI for these predictions were obtained by bootstrapping , following the exact same procedure as for the CI for mu ( described above ) .
The data consisted of 9 time series observations for the period from 1990 to 2011 , during which the snail population density ranged from 300 to 2108 m−2 ( Table 1 ) . The data were first subjected to a simple linear regression ( SLR ) . The SLR gave no indication that the slope of the regression was significantly different from zero ( P = 0 . 7959 ) ; however , the F ratio ( 0 . 0722 ) was not significant ( Table 2 ) , suggesting that the SLR might not be a suitable model for these data . In addition , plots of residual errors against their fitted values were not random , QQ-plots were not normal and leverage plots showed large Cook's distances for the 1991 and 1995 observations ( see Figure S3 ) . The critical Cook's distance for these data was 0 . 444; this was exceeded by the 1991 observation ( distance almost 1 . 0 ) and approached by the 1995 observation ( distance ∼0 . 40 ) . Consequently , the 1991 observation was classed as an outlier , with the 1995 datum retained as a borderline observation . The 1991 observation was unusually low ( 300 m−2 ) and the 1995 observation unusually high ( 2108 m−2 ) , with the mean population density being 1177 m−2 . Table 1 shows the effect of removal of the outlier on the mean and standard deviation ( SD ) . A Poisson general linear model ( GLM ) was also fitted to the data . The result was highly significant ( Table 2 ) , indicating that the GLM was , like the SLR , not a good model for describing these data . The quotient of residual variance over degrees of freedom for the Poisson GLM was 237 and 69 . 65 ( for the full data set and without outlier , respectively ) , suggesting that the counts were over dispersed . In view of this the negative bionomial GLM , another Poisson-based model , was applied and found to be a better fit to the data than either SLR or Poisson GLM ( Table 2 ) . Unlike the Poisson GLM , the residual deviance for the negative binomial was not significant ( P = 0 . 232 and 0 . 237 , full data set and without outlier respectively ) . The negative binomial for the data without the outlier indicated a significant slope to the trend ( −0 . 0004606 ) ; however , the slope for the full data set was not significantly different from zero ( P = 0 . 771 ) . In view of the fact that the 1991 observation was clearly an outlier and that there was no a priori hypothesis to predict the low population density in that year , the analyses proceeded with both the full data set and one with the 1991 observation omitted . The SLR for the partial data set was also not significant ( but showed a slightly better fit to the data partial data set than to the full set ) and , like the full data set , a negative slope was indicated ( Table 2 ) . The GLM was also rejected for the partial data set . The Shapiro-Wilke Normality Test for the full data set suggested that the data might be normally distributed ( W = 0 . 9162 , P = 0 . 3615 ) ; however , that for the partial data set indicated that the data were not normally distributed ( W = 0 . 7909 , P = 0 . 02286 ) . Independent runs of REML searches ( coupled to SAN ) were found to converge on the same parameter estimates and REML values , regardless of the starting parameters ( which were either the output of a run of ML searches ( each based on the output of the previous search ) or random variants around the LLR and SEG estimates ) . Also , improvement in restricted likelihood was found after 10 runs of 1 million iterations each ( with different starting parameter values ) . Consequently , it is likely that the parameter estimates reported ( Table 3 ) are the best estimates or close to them . Plots of log likelihood against mu also suggested that maximum likelihood estimates had been found ( Figure S4 ) . As seen in Table 3 , after fitting the modified Gompertz linear state space model ( GSS ) , the REML estimate of the trend parameter ( mu ) for the full data set was positive , whilst that for the partial data set was negative , although the 95% confidence intervals ( CI ) for the two estimates overlapped one another ( Table 4 ) ; the mu estimate in both cases was very small . The REML for the full data set fitted to the GSS was −6 . 464139 and that for the partial data set was −16 . 83423 . In the case of the full data set , the empirical ( bootstrapped ) CI completely overlapped the equivalence region , but exceeded it at both ends . The empirical mu ( 1 . 52350e-04 ) lay well inside the upper limit of the equivalence region ( CI for the simulations with mu fixed at zero ( see Figure S5 ) ) . From this it is reasonable to conclude that the true mu might be >0 but is certainly close to or equal to zero . In contrast , with the partial data set , the CI for the empirical mu lay entirely within the equivalence region; therefore the hypothesis that mu is significantly different from zero may be rejected at P<0 . 05 . Using the REML parameter estimates , and under the GSS model , the Kalman filter was employed to predict the true population density ( Xt ) for both the stationary and non-stationary cases ( Table 5 ) . The predictions and the observed ( original ) data ( Yt ) are plotted in Figure 2 . The Kalman filter was also used to predict values for key years and CIs for these predictions obtained by bootstrapping the data 1 million times . For these predictions only the partial data set was used because the 1991 observation was unlikely to have been a reflection of natural events or consistent with long-term environmental processes . Similar predictions were also made using the SLR and GLM models , where statistically appropriate . The predicted values from the negative binomial GLM were very close to those of the Poisson GLM and so only the values for the negative binomial are given in Table 5 ( as the negative binomial was the statistically preferred model of the two ) . The similar predictions from the two GLM models is not unexpected because the negative binomial mainly differs from the Poisson GLM in the way in which the likelihood function penalizes class values with high variances , and this difference may not always be sufficient to be reflected in parameter estimates; however , it is likely to affect their variance [30] . The trend was extrapolated and the snail population density in 2020 predicted . This prediction is based on observed densities up to and including the 2011 observation , and assumes the same growth trend operates from 2012 to 2020 as operated between 1990 and 2011 . The population density at Ban Hin Laht in 2020 was predicted to be 1216 . 884±244 . 0057 which is very close to , and not significantly different from , the average population density observed today ( P<0 . 05 ) . It should be noted , however , that the predicted values are estimates of the true population density ( Xt ) rather than the expected observed density ( Yt ) . In contrast , the SLR and GLM models predicted that population densities would be much lower in 2020 than at present . The 1991 population density was also predicted using the GSS and estimated at 1408 . 522±493 . 2719 ( Table 5 ) ; this is significantly greater than the observed density of 300 and confirms that the 1991 observation was atypical and must reflect some event occurring at Ban Hin Laht prior to sampling . Predictions were also made for 1995 and 2002 , years for which data were available; however , these are key years because 1995 is the first year after dam operation and the 2002 sample was taken after the dam had been kept open for almost one year . In these cases initiating the Kalman filter under non-stationary or stationary assumptions had a small effect on the predictions . The 1995 observation ( 2108 m−2 ) was significantly greater than the predicted value ( 1270 . 391±302 . 1927 ) and this was also found with the SLR and GLM models . In contrast , the 2002 observation was less than the predicted density ( 979 m−2 cp . 1295 . 712±588 . 5236 m−2 ) ; however , this was not statistically significant with any model , except for the negative binomial GLM which predicted a value significantly lower than the observed density ( Table 5 ) .
No significant large population growth trend was indicated at Ban Hin Laht , either with the full or partial data set . Under the GSS model the partial data set indicated that the population growth parameter was not significantly different from zero . Consequently , the data support the view that population densities have remained fairly constant over the sampling period . Furthermore , predictions of the 2020 population density , based on parameters estimated using the GSS model , were that the density in 2020 will not be significantly different from the average density observed between 1990 and 2011 . The observed population density for 1991 was less than a quarter of the value predicted using any model ( SLR , GLM or GSS ) . This observation was taken in the first dry season after construction of the Pak-Mun dam began ( see Table 6 ) . It is possible that silt , debris or toxic materials could have been washed from the construction site and affected the viability of N . aperta eggs deposited at Ban Hin Laht . The study site is only 33 km from the confluence with the Mekong river and flow reversal could occur at Ban Hin Laht at the height of the flood period in the Mekong river . Indeed , Mekong river backflow 70–100 km upstream from the confluence has been predicted for the Nam Songkhram river [31] . Further , N . aperta is known to be sensitive to depositing silt and prefers a fine eroding substratum [12] . An alternative explanation is that there was some undocumented attempt to control the snails ( probably using molluscicides ) after local villagers began to protest over the risk of schistosomiasis . By the following April the N . aperta population density recovered to levels slightly exceeding the study average , suggesting that , if the dam construction had effected the drop in density , the impact was associated with the early stages of construction ( most likely the initial movements of earth and excavations ) . The 1995 observation was taken in the second dry season following completion of the dam ( Table 6 ) , and is the time of the highest population density observed in this study . The observed density was almost double ( and statistically significantly greater than ) the predicted values with either the SLR , GLM or GSS models , and so indicates a marked departure from the overall population trend running through the study period . Between 1992 and1995 the Pak-Mun dam effected a shift toward slightly more lentic conditions at the study site ( a reduction in flow rate and increase in depth [9] ) . It is conceivable that reduced flow rates and delaying of the onset of the annual flood could increase survival and breeding success in N . aperta and this may have led to the increase in population density observed; however , by 2002 densities had fallen to levels below average and below GSS and SLR predictions ( although not significantly ) . The high densities in 1995 could be explained by a process similar to that observed shortly after filling of artificial water impoundments , where fish stocks of many indigenous species rise in the short-term before declining in the longer term; this is due to factors such as expansion into new ecological niches ( that are not the usual niche of the species concerned ) in the absence of competitors and barriers such as high current [32] . The 1995 observation might also be explained as increased sampling efficiency due to ease of diving in slower flowing waters than those usually encountered when making observations at Ban Hin Laht; however , this implies that densities recorded during 2002–2011 should also be high , unless some overriding factor were acting to reduce snail populations during those years . The fluctuations in observed density might be explained by sampling bias due to population growth or decline across the sampling period ( which ranged from April 7 to May 10 ) . During this time interval , which lies well within the low water ( dry season ) period , snail populations in the Mun River have not been observed to increase in density and are quite stable , but the snails do increase in shell size . This could slightly increase sampling efficiency in May relative to April ( because larger snails might be easier to remove from the stones ) ; however , initial investigations into the design of the sampling procedure suggested that even juvenile snails were removed by the technique used in this study . In contrast the increase in shell size by May could effect a lower on-stone snail density because fewer snails could be accommodated on the same sized stone , although at times of very high observed densities the snails were found to climb on top of one another and graze the shells of other snails in the same manner as they would graze the stone itself – even though suitable areas of the stone's surface were still not yet fully occupied . In any case , both high and low densities were recorded in April and May of different years , so that there did not appear to be any trend correlated with sampling month . For example , the highest and the lowest observed densities both occurred in early April . The 2002 observation was much lower than the density predicted by any model; however , this was not statistically significant with any model except the GLM . This fall in density was observed after the dam operation had been changed to keep the gates open ( allowing near normal flow of the Mun river ) for the preceding two years; thus suggesting that keeping the dam open might lower snail population densities . It is tempting to attribute this fall to the year round opening of the dam , especially as the population density returns to that of normal dam operation levels after the policy change to gate closure during the flood period . Opening of the dam could limit snail population growth by allowing the natural onset of the spate , at least one month earlier than when the gates are closed , which terminates the breeding period of N . aperta . The fall in density was , however , not statistically significant and there are many other factors which could explain the decline ( for example , natural cycles in N . aperta population density or that of competitor species , local pollution from sources unrelated to Pak-Mun and increases in Mekong river flood volume ) . In conclusion , this study has shown that there has been no marked increase in N . aperta population growth following operation of the Pak-Mun dam and long-term projections are that over the ten years following the study snail population densities will remain close to their current average . Indeed , the growth trend parameter estimated for the time series observations was not statistically different from zero , implying that N . aperta populations near Pak-Mun were quite stable . Nevertheless , the analysis does indicate that changes in snail population density did co-occur with changes in operation of the dam , although only two out of three of these changes were statistically significant ( i . e . , 1991 , 1995 ) . The present study is , however , limited in its scope and the applicability of its findings . The study examined population changes at only one site on the Mun river and changes observed might be peculiar to Ban Hin Laht . Similarly , no samples were taken downstream of Pak-Mun where the ecological impacts of the dam are expected to be quite different . Confidence in the conclusions of studies of this kind will be greatly improved by further work at additional sites around the dam ( if data are available ) and by extending the present data set with further , regular , observations beyond 2011 . In view of the fact that many dams are now planned across the range of N . aperta and S . mekongi , it is vital that we better understand the effects of dams such as Pak-Mun on snail population growth . The present work suggests that studies of this kind are useful in understanding the impacts of water impoundment but data are currently severely lacking . Investment in research into this area is urgently required to enable the planning of public health compliant water resource development in the lower Mekong Basin . | There is much controversy over the effects of water resource development on the transmission of schistosomiasis in the lower Mekong Basin . Impact assessments are urgently required because there are currently 12 such projects planned in the region . The key to understanding the effects of impoundment is the impact on the snail intermediate host , which , in the case of Mekong schistosomiasis , is Neotricula aperta . Surprisingly , we have almost no data on N . aperta population trends nor on the impact of dams . To address this , the present work focused on a population near the Pak-Mun dam in Thailand . The analysis suggested that N . aperta populations were not growing significantly over the study period ( 1990–2011 ) , but that the dam may have affected a spike in population density immediately after its completion . The study also revealed changes in density that were coincident on changes in operation of the dam; suggesting that keeping the dam open might lower snail population densities . This is the first scientific assessment of the impact of the Pak-Mun dam on N . aperta and suggests that dams of this kind may affect snail population density . The study also indicates an urgent need for additional independent observations and continuing regular surveys . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [] | 2013 | A Population Growth Trend Analysis for Neotricula aperta, the Snail Intermediate Host of Schistosoma mekongi, after Construction of the Pak-Mun Dam |
Recognition of individuals by scent is widespread across animal taxa . Though animals can often discriminate chemical blends based on many compounds , recent work shows that specific protein pheromones are necessary and sufficient for individual recognition via scent marks in mice . The genetic nature of individuality in scent marks ( e . g . coding versus regulatory variation ) and the evolutionary processes that maintain diversity are poorly understood . The individual signatures in scent marks of house mice are the protein products of a group of highly similar paralogs in the major urinary protein ( Mup ) gene family . Using the offspring of wild-caught mice , we examine individuality in the major urinary protein ( MUP ) scent marks at the DNA , RNA and protein levels . We show that individuality arises through a combination of variation at amino acid coding sites and differential transcription of central Mup genes across individuals , and we identify eSNPs in promoters . There is no evidence of post-transcriptional processes influencing phenotypic diversity as transcripts accurately predict the relative abundance of proteins in urine samples . The match between transcripts and urine samples taken six months earlier also emphasizes that the proportional relationships across central MUP isoforms in urine is stable . Balancing selection maintains coding variants at moderate frequencies , though pheromone diversity appears limited by interactions with vomeronasal receptors . We find that differential transcription of the central Mup paralogs within and between individuals significantly increases the individuality of pheromone blends . Balancing selection on gene regulation allows for increased individuality via combinatorial diversity in a limited number of pheromones .
Animals produce complex blends of scents that can provide information on species , age , sex and individual identity [1] . As arguably the most complex and subtle form of recognition , individual recognition has attracted considerable attention from the perspective of neurobiology [2–4] and chemical ecology [5–7] . To be useful for individual recognition , scents must be variable among individuals in a population but stable within a particular individual over time [8–10] . Using discrimination tests , researchers have documented that a wide range of species are capable of differentiating between scents based on variation in multiple components of complex mixes of chemicals [8 , 11] , leading to the early proposition that a wide range of genetic differences among individuals contributed to individual recognition [10 , 12–14] . Indeed many species produce consistent individual odor profiles that have the potential to mediate individual recognition [10 , 12–14] , though the components of odors that are biologically relevant for individual recognition are generally unknown [15 , 16] . Recent studies using behaviorally salient tests of individual recognition rather than discrimination have identified specialized semiochemicals directly encoded in the genome as necessary and sufficient for individual recognition via scent marks in house mice [3 , 5 , 17] . These findings raise two important questions . First , what is the nature of genetic variation underlying individually distinctive odor signatures ? A key feature of individually distinctive scent profiles is their combinatorial nature , which could arise through a range of mechanisms including variation in coding sequences , gene transcription and protein translation . Determining the genotype-phenotype map has a number of important implications . The potential for pleiotropic effects varies between different elements of coding and non-coding DNA , which is expected to influence the rate of adaptive evolution [18] . Understanding the mechanisms giving rise to a phenotype are important for interpreting patterns of molecular evolution [19] . Finally , the relationships between chemical cues and genetic variation are widely discussed in the literature [12 , 20–22] , but the mechanisms linking genotype and phenotype are generally unknown . Elucidating the nature of genetic variation underlying individually distinctive traits will provide insight into how well phenotypic differences could be used to infer genetic differences among individuals in a population . Second , how is the diversity in specialized semiochemicals maintained ? Models where individual recognition is mediated by a wide range of genes and metabolic processes propose that identity cue diversity is derived from neutral genetic diversity [23] or maintained by balancing selection on loci for other reasons , such as immune selection on MHC [24] . Thus , individually distinctive scent signatures have often been seen as identity cues rather than signals selected to advertise individual identity [14 , 24] . A separate line of research , however , argues that elevated diversity in phenotypes used for individual recognition is maintained because identity advertisement is favored when confusion among individuals results in costly misdirected behaviors , such as unnecessary aggression or missed mating opportunities [25–28] . Rare phenotypes are favored by negative frequency-dependent selection leading to the elevated diversity in identity signals [28 , 29] . Indeed , comparative phenotypic studies provide evidence that selection for identity signaling maintains variation in appearance [30] , scent [31] or vocalizations [32 , 33] depending on the species . While selection for individual identity signals is expected to maintain increased diversity at relevant loci [34] , molecular evidence of frequency-dependent selection on specialized semiochemicals mediating individual recognition in animals is lacking [24] . In this study we examine the nature of genetic variation underlying individually distinctive urinary protein pheromones in a wild population of house mice and test for patterns of selection . The individual signature in mouse urine comes from combinatorial differences in the presence and relative abundance of isoforms of major urinary proteins ( MUPs ) ( Fig 1A [3 , 5 , 35 , 36] ) . MUPs are detected by the vomeronasal organ and they also bind volatile pheromones that influence scent . The combinatorial protein variation mediates countermarking behavior in males as well as sexual preferences in females [3 , 37] . Previous work has demonstrated that MUP profiles are genetically determined and stable within adult individuals , and are highly heritable [5 , 17 , 38–40] . MUPs are the products of a family of tandemly-arrayed Mup genes and associated pseudogenes on chromosome 4 , with more than 20 coding genes ( Fig 1B and 1C [41–43] ) . The genes are broken into two categories based on their phylogenetic relatedness and sequence similarity , termed peripheral and central Mups corresponding to their position along the chromosome ( Fig 1B ) . Variation in the excretion of proteins produced by the central Mup genes , which are >98% similar in coding sequence , is the main basis of individual variation in MUP phenotype in house mouse urine [3 , 41] . The nature of variation in central Mup genes among individuals in wild populations that gives rise to individuality of urine markings has yet to be determined . Categorizing and analyzing molecular variation in dozens of nearly identical paralogs in natural populations presents a substantial challenge [44–46] . In the present work we sequenced large amplicons ( ~5kb ) and developed approaches for analyzing patterns of variation among the highly similar central Mup gene complements across individuals . Though we are not able to assemble full gene sequences , our approach allows for tests of adaptive molecular evolution by comparing patterns of diversity among many different types of sites ( e . g . non-synonymous , synonymous , etc . ) . The present study seeks to determine the nature of genetic variation and assess what role , if any , selection on Mup genes has played in promoting and maintaining individuality in protein pheromone blends in wild house mouse urine . Using the F1 male offspring of unrelated wild-caught parents from one population ( S1 Table ) , we dissect the contributions of variation in coding sequences , gene expression and protein translation to individuality in the urinary identity signals of house mice . We confirm a role for both differences in coding sequence and regulatory variation among the central Mup paralogs in generating individually distinctive MUP blends . Importantly , our work demonstrates that differential transcription of central Mup genes substantially increases differences in pheromone blends among individuals in a wild population . Analyses of sequence data provide evidence of frequency-dependent selection acting on both coding and 5’ regulatory sequences of the central Mup genes , with a particularly strong signal of selection on regulatory sequences . Taken together , these data highlight the importance of selection on gene expression in shaping combinatorial variation in pheromone blends used for individual recognition .
We measured the relative abundance of MUP isoforms at eight distinct molecular weights in the urine of 18 male laboratory-born descendants of wild mice ( each individual from different wild-caught parents ) using electrospray ionization mass spectrometry ( ESI/MS ) . We collected and analyzed two urine samples from each sexually mature mouse ( average age at time of urine collection = 218 days , range 190–251 days ) , which had all previously mated ( S2 Table ) . The urine of mice in our sample shows the two hallmarks of an identity signal [5 , 47] . First , there is heterogeneity in the urinary protein profiles among mice; they differ in the presence as well as relative abundance of protein isoforms in mass spectra ( Fig 1A ) . Second , each mouse produces a consistent pattern of urinary proteins ( median Pearson correlation coefficient between relative abundance of protein masses in two independent samples from an individual: 0 . 98 , range: 0 . 90–0 . 99 ) . The urine samples analyzed from each individual were collected from 0–30 days apart , though there was no association between the time between urine samples and the similarity of those samples ( S1A Fig , linear model , t16 = 1 . 45 , r2 = 0 . 12 , P = 0 . 17 ) demonstrating that the relative proportion of isoforms excreted in the urine is stable over time . Pairwise comparisons among the profiles of different mice show significantly lower correlations than between two different samples from the same mouse ( S1B Fig , Kolmogorov-Smirnov test , D = 0 . 55 , P< 0 . 0001 , median Pearson correlation coefficient between samples from different individuals: 0 . 80 , range: -0 . 08–0 . 99 ) . Multiple MUPs with similar molecular weights ( within the instrument capability of ±1–2 Da ) are indistinguishable using the approach reported here and thus , variation between individuals is potentially underestimated using the ESI/MS methods of this study [48] . We used primers in conserved regions flanking the 5’ and 3’ ends of genes to amplify the central Mup genes , producing an amplicons of approximately 5Kb ( Fig 1C ) . The high sequence similarity among the central Mup paralogs [41 , 42] prevented us from assembling full gene sequences or unambiguously aligning these reads to the reference genome . We took advantage of the high sequence similarity to develop a novel technique for analyzing sequence reads from the weakly diverged paralogs of multigene families ( Fig 2A ) . Because the paralogs are so similar , reads from any one paralog readily align to all the others . Rather than aligning the reads to all of the central Mup genes found in the reference genome , we created a reference alignment file containing only one central Mup gene . Here we report data for the alignment of reads to Mup11 ( nomenclature from the MGI database , [49] ) , although alignments to other central Mup genes produce the same results . The resulting alignment to a single central Mup gene provides information on the presence and relative abundance of mutations among all the paralogs amplified from an individual . This approach does not distinguish between allelic and paralogous variation , but it does provide a summary of all the Mup variation among mice that may contribute to individuality in MUP profiles . Hereafter , we simply refer to the combination of allelic and paralogous sequence differences as variants . The signal peptide at the N-terminal of the protein is cleaved prior to excretion in the urine [50 , 51] , so we focus our results on the mature gene product that is excreted and can contribute to individuality in scent ( Fig 1D and 1E ) . Our data provide clear evidence for heterogeneity in coding sequences contributing to individuality of urine in wild house mouse populations . We find evidence for 11 segregating non-synonymous variants in the mature proteins , three of which are not found among the central Mup paralogs in the mouse reference genome ( GRCm38[41] ) ( Table 1 ) . Three of the eleven variants are present in all the samples though occur at variable rates among individuals ( Table 1 ) . Variation in the percentage of non-reference reads may arise from the combination of a variable number of paralogs containing the mutation among individuals , allelic variation among paralogs , and variation in the total copy number of Mup paralogs among individuals . Our sequencing and alignment approach is not able to readily distinguish between these causes . The other 8 mutations are present in a portion of the individuals and tend to occur at low levels when they are present in fewer individuals ( linear regression , r2 = 0 . 34 , t = -2 . 15 , P = 0 . 06 ) , suggesting that rarer variants among individuals also tend to occur in few gene copies within individuals . All but one non-synonymous variant is shared by at least two individuals , though that variant occurs at reasonably high frequency in the one individual ( 19% of reads , Table 1 ) . Collectively these data suggest that the 11 non-synonymous variants detected do not include any false positives . Furthermore , none of the variants detected are predicted to produce premature stop codons or frameshifts ( Table 1 ) , as would be expected if our alignment had contained sequences from Mup pseudogenes . All of the amino acid variants identified can be mapped to the structure of MUP 11 ( 1I04 . PDB , [52] ) ( S2 Fig ) . With the exception of the frequently observed Phe->Val mutation at position 56 ( mature protein numbering scheme ) , all of the amino acid substitutions are located on the surface of the protein , with one set of changes being located in a ‘patch’ at the position of two external loops . These changes are consistent with interactions with receptor molecules , and to some degree , with alterations in cavity volume and potentially , specificity of binding . Using identical alignment and variant calling methods , we found evidence for 22 synonymous variants , of which nine were found only in a single sample with a relatively low percentage of reads ( Table 2 ) . Eight of these nine variants were from one mouse ( individual TAS293 ) and were present at around 1% . These low frequency variants may be rare alleles or false positives . The particular mouse in question had higher than average read coverage ( >10 , 000x ) so it is possible that the ~1% non-reference reads are attributable to sequencing error which surpassed our cutoff ( see Methods ) . Removing this individual ( TAS293 ) from subsequent analyses of sequence variation related to selection did not affect the results . While it is possible that some of the synonymous mutations reported here are false positives , that synonymous mutations are at much lower frequencies than non-synonymous mutations is unambiguous in this dataset . Next , we sequenced liver transcriptomes , where urine-excreted MUPs are produced [53 , 54] , and compared patterns of variants between DNA amplicons and the Mup transcripts to assess the role of gene expression in individual identity . To ensure deep sequencing of Mup transcripts , we size-selected cDNA libraries made from liver RNA to over-enrich Mup transcripts relative to the rest of the liver transcriptome . We followed the same approach used to align DNA , and we aligned RNA reads against all peripheral Mup genes with only Mup11 representing the central genes . Comparisons of RNA and DNA demonstrate differential rates of transcription among Mup sequences within and between mice . While it was possible to assemble long transcripts , our inability to generate full assemblies of DNA amplicons precludes direct comparisons of DNA and RNA across the entire length of the transcript . However , many reads span the full length of exons in both the RNA and DNA datasets; we compared abundances of unique reads of exon 1 , the most variable exon , between the two datasets to assess patterns of transcription . Our data demonstrate that some of the DNA sequences identified in each mouse are not transcribed ( Fig 3A ) . The lack of transcription raises the possibility that the sequences detected in the DNA data are actually pseudogenes , although two lines of evidence argue against this: ( i ) none of the variants in the DNA pool are early stop codons or frameshift mutations ( Table 1 ) and ( ii ) the same sequences that are not transcribed in one individual sometimes show evidence of transcription in others ( Fig 3A ) . Furthermore , previous studies of inbred lab strains have also demonstrated that mice do not transcribe all of their central Mup genes [41] . Overall , we find a weak positive relationship between the relative abundance of reads in DNA and RNA for exon 1 ( linear regression , r2 = 0 . 26 , t207 = 8 . 43 , P < 0 . 0001 ) . Even if DNA sequences that are not expressed ( i . e . , represent < 1% of RNA reads in a given individual ) are excluded , the relationship remains modest . ( Fig 3B , linear regression r2 = 0 . 25 , t101 = 5 . 86 , P < 0 . 0001 ) . The positive relationship suggests that DNA sequences that occur in more copies tend to occupy a larger share of the mRNA pool . To assess the extent to which post-transcriptional processes such as variable translation or decay of proteins influence MUP phenotypes , we compared the predicted proportional abundance of MUP masses based on our mRNA data ( S3 Fig ) with the proportional abundance of urinary MUP masses measured by ESI/MS . The mRNA data very accurately predict the observed amount of each mass peak in a mouse’s urinary MUP profile ( Fig 3C , linear regression , r2 = 0 . 91 , t73 = 27 . 48 , B = 0 . 98 , P < 0 . 001 ) , suggesting little if any role for variation in post-transcriptional processes in contributing to phenotypic differences among individuals . The tight relation between patterns of mRNA and urinary proteins , despite an extended period between collection of the urine and mRNA samples ( average = 54 . 2 days , range 9–212 days ) , emphasizes the stability of the relative expression patterns of different MUP isoforms in adult mice and highlights their potential as an individual identity signal . Indeed , a statistical model explicitly incorporating time between urine and RNA collection shows that the tight relationship between the proportional abundances of mRNA and observed proteins does not change over time ( P = 0 . 33 ) . Furthermore , the ability of our method to analyze RNA to robustly predict the observed urinary protein patterns provides validation for the alignment methods employed in this study . Differential silencing of some Mup paralogs could potentially be achieved through a diverse set of mechanisms , though variation in cis-regulatory regions seems likely because the relative expression of urinary protein isoforms is stable even as individuals alter their levels of total urinary protein output [5] . We examined a ~325 bp region including the first exon and upstream sequences for evidence of sequence variation associated with expression . Based on stringent criterion ( see Methods ) we classified promoters as expressed ( N = 35 ) or silenced ( N = 97 ) . Using this restrictive dataset , we identified 10 eSNPs ( Fisher’s exact test , P < 0 . 05 ) as being significantly associated with differences in expression category ( Figs 4A and S3 ) . We note that more variants are found only in the silenced category and also likely influence expression but are not common enough among the silenced promoters to register as significant given our restrictive dataset . Whether the mutations immediately upstream of the gene are causative is unclear , nevertheless they provide a potential mechanism for differential expression of Mup loci . Despite silencing of some genes we do not detect frameshift mutations or early stop codons in the data set ( Table 1 ) , suggesting that although the upstream regions have become non-functional the exons have been maintained in a functional state . Comparisons of neighbor-joining trees for the promoter and exon 1 sequences demonstrate that a single exon 1 sequence may be associated with a diversity of promoter sequences across a population and within a single individual , consistent with a history of non-allelic homologous recombination ( Figs 4B and S5 ) . Variable gene expression could make individual scents more similar if shared sequences are expressed while rarer sequences are disproportionately silenced or under-expressed . Conversely , the preferential silencing or under-expression of common sequences would give disproportionate weight to rarer sequences resulting in greater phenotypic differences among individuals . To assess the extent to which variable gene expression either increases or decreases phenotypic differences among individuals , we compared the extent of differences between the same individuals at the DNA and RNA level . We calculated a pairwise differentiation index , PD , for all possible pairings of individuals by summing the differences in the percentage of non-references reads between individuals ( Fig 2B and 2C ) . While PD ignores the linkage among variants ( which we could not easily determine for DNA ) , it provides a reasonable proxy for the level of differences in urinary protein phenotypes based on DNA and RNA , especially when limited to non-synonymous variants that might lead to detectable scent differences among mice . While selection analyses should use values of PD corrected for the number of sites , we note that raw , uncorrected values of PD reflect total differences between two individuals and are appropriate for comparisons of predicted phenotypes based on DNA and RNA , especially given that the sequences compared are of the same length . To differentiate between the two uses of the index we define PD’ as the total difference between two individuals and PD as the difference between individuals corrected for the number of sites compared . Variable gene expression among individuals increases the differences between pairs of mice in a disproportionate number of pairwise comparisons ( Fig 5A , Mann-mean RNA PD’–DNA PD’ = +0 . 47 , one-sample t test , t152 = 7 . 70 , P < 0 . 0001 ) . Though some pairs of mice end up more similar as a result of variable gene expression , in general mice are more distinctive at the RNA than DNA level . From the perspective of individual mice , we see an increase in average PD’ for 16 of the 18 individuals as a result of variable expression of central Mup genes among individuals ( Fig 5B , t17 = -4 . 65 , P = 0 . 0002 ) . Both coding and regulatory variation contribute to individuality , though it is unclear from our previous analyses whether the variation observed is consistent with neutral evolution of the Mup gene family or the result of selection for distinctiveness . To look for evidence of selection we compared patterns of variation at synonymous and non-synonymous sites . Though we do not know the precise number of gene copies measured across all individuals in our sample , it is constant across all the sites considered so variation in the percent of reads with a variant for each individual can be considered an estimate of the relative number of gene copies with a particular variant . Overall , we find higher diversity in synonymous sites compared to non-synonymous sites ( Fig 6A , median PDNon-syn = 0 . 00127 , median PDSyn = 0 . 00328 , MWU , U = 4877 , n = 153 pairwise comparisons , P < 0 . 0001 ) , consistent with high levels of purifying selection previously detected among the central Mup paralogs [55] . Though the overall coding sequence is under purifying selection , two observations suggest that non-synonymous variants may be under frequency-dependent selection . Non-synonymous variants are present in more individuals ( Fig 6B , MWU , U = 39 , P < 0 . 002 ) and occur at higher frequencies within individuals compared to synonymous mutations ( Fig 6C , MWU , U = 23 , P = 0 . 0002 ) . To investigate selection more formally , we conducted individual-based forward-time simulations of a single model of frequency-dependent selection using realistic demographic parameters . We then compared simulated numbers and frequencies of non-synonymous and synonymous mutations with observed values using approximate Bayesian computation [56] . The results of the simulations demonstrate that the patterns of variation described here are consistent with a history of relatively strong frequency-dependent selection maintaining diversity in central Mup genes at segregating replacement substitutions ( Fig 6D ) . Importantly , the 95% credible interval for the strength of selection on MUP individuality signaling does not include zero ( see Methods ) . Therefore , it is unlikely that neutral process could have generated the patterns of variation observed in the dataset . As with any simulation , the framework described here represents an approximation of the evolutionary processes that generated these patterns of genetic diversity in the central Mups . Nonetheless , these simulations provide support for the assertion that patterns of nucleotide variation are consistent with frequency-dependent selection maintaining non-synonymous diversity at select sites in an otherwise conserved gene sequence . We also assessed the level of variation in the peripheral Mup genes , for which there is no current evidence of individual variability . Three of the genes sequenced , Mup4 , Mup5 and Mup6 , are not expressed in the liver and not excreted in the urine [41] . Two of the peripheral genes , Mup3 and darcin ( Mup20 ) , are expressed in the liver of males and excreted in the urine but serve separate signaling functions apart from individual identity . In males , both Mup3 and Mup20 elicit aggression [3] . Mup20 mediates learned place preferences for the sites of previously encountered scent marks in both sexes [57] together with learned attraction to airborne scent from an individual male’s scent marks [58] and substantially enhanced neurogenesis in the hippocampus of females [59] . For three of the genes–Mup3 , Mup4 and Mup5 –we did not identify any exonic variants in the population . We found two non-synonymous variants in Mup6 and one in Mup20 , all segregating at low frequencies ( S3 Table ) . These observations provide no evidence for frequency-dependent selection on the peripheral Mup genes–indeed the lack of variation is indicative of purifying selection on the peripheral Mup genes . Instead , selection that maintains sequence variation appears limited to the central Mup genes involved in producing individually distinctive urinary scent marks . While the increase in PD’ between DNA and RNA is consistent with selection acting on regulatory sequences to increase individuality ( Fig 5 ) , it is possible that increased diversity merely results from the process of random mutations that silence genes or the duplication of already silenced genes . If this were the case , then we expect to see a similar increase between DNA and RNA in the synonymous differences among individuals , PD’Syn . Counter to the null prediction , we find a decrease in PD’Syn in RNA compared to DNA ( Fig 7A , MWU , U = 15783 . 5 , P < 0 . 0001 ) . We also compared the diversity of promoter sequences to introns from the central genes . Variants in the ~200 bp region upstream of the start codon tend to be found in similar numbers of individuals as variants in the introns ( Fig 7B , MWU , U = 8434 . 5 , n = 55 promoter variants , 270 intronic variants , P = 0 . 11 ) . However , promoter variants are found at higher frequencies compared to intronic variations ( Fig 7C , MWU , U = 8905 , P = 0 . 0198 ) , consistent with selection maintaining diversity in promoter sequences . PD is greater in the promoter region compared to the introns ( Fig 7D , MWU , U = 5333 , n = 153 pairwise comparisons , P < 0 . 0001 ) . This is consistent with diversity-enhancing selection on promoters or with lower constraint on promoters , although we note that introns are generally less constrained than 5’ flanking regions [60] . As with coding sequence diversity , comparisons between the central and peripheral Mup gene promoter sequences can determine whether the molecular patterns observed are specific to the central Mup genes involved in individual identity . Extensive promoter variation among the central Mup genes is unusual compared to the peripheral Mup genes , where we identified very little variation in promoter sequences ( S3 Table ) . In both Mup4 and Mup6 we found no variants in the paralogous section of promoter sequence examined in the central Mup genes . We found one variant each in Mup20 and Mup5 segregating at low frequency ( S3 Table ) .
Dissection of the genetic basis of individuality in the urinary scent marks of house mice demonstrates a role of selection on both coding and regulatory variation in maintaining adaptive identity-signaling phenotypic diversity . Our results confirm previously proposed roles for variation in gene sequences , the number of gene copies and transcriptional regulation in determining pheromone blends [41] . DNA sequences are weakly correlated with transcript abundance due to variable silencing of genes associated with mutations in promoter sequences ( Figs 3 and 4 ) . Whereas mRNA and protein abundances in general tend to be imperfectly correlated [61] , we find a robust relationship between liver mRNA and proteins excreted in the urine across mice in our sample ( Fig 3C ) . This is unsurprising , as MUPs are rapidly excreted from liver and there is no opportunity for intracellular degradation that can compromise the mRNA:protein ratio . This means that analysis of urinary MUP output is an accurate reflection of hepatic translational capacity , with the caveat that some distinct transcripts direct the synthesis of proteins of the same mature mass . A number of studies have shown that the total amount of MUPs excreted in urine can vary over the course of an individual’s life [62–66] and between generations [67] . Thus , there is abundant evidence that the total amount of MUP excreted responds to social and physical conditions . However , the total amount of MUP does not provide identity information , in the same way that the volume of a voice does not identify a speaker . Thus , the critical question with regard to discrimination and recognition is whether the pattern comprising the presence and relative abundance of different MUP isoforms is consistent within individuals . The stability of MUP patterns for a given genotype is evidenced by the fact that individuals that share MUP haplotypes produce the same relative pattern ( though not necessarily absolute amount ) of MUP isoforms in their urine as adults [5 , 17 , 40 , 68 , 69] . A recent study using isoelectric focusing ( IEF ) to identify MUP patterns suggested that individual MUP profiles change over the course of adulthood [39] . However , the Thoß et al . study [ref 39] was based on unidentified gel bands that were not assigned to any gene products , confounding central MUPs , peripheral MUPs and any other urinary proteins of similar isoelectric point . Further , many of these bands were labeled as “minor” and it is likely that some were generated by artefactual changes caused by storage , notably deamidation of asparagine residues [70 , 71] , an observation that caused us to abandon IEF as a method of profiling MUPs in all but freshly collected urine samples . Nonetheless , the Thoß et al . [ref 39] study still found a very high degree of consistency in bands expressed within individuals ( median similarity of MUP profiles 94% , Figure 4 in ref [39] ) , confirming the long-term stability of the MUP pattern . Using genomic and ESI/MS methods , we provide clear evidence that proportional abundances of central MUP isoforms and corresponding liver Mup mRNA are very tightly correlated , even when measured more than six months apart ( Fig 3C ) . Stable expression of identity signals is necessary for recognition , though identity signals need not be immutable to provide this function–for example , human faces have evolved to signal individual identity [34] and change to some extent with age , notably at sexual maturity . Electrophoretic studies show that expression of MUP isoforms develops progressively during adolescence [64 , 72] , though they stabilize during adulthood [72] . Indeed , our transcriptomic data ( and the tightness of the transcript:protein relationship ) suggest that the proportional abundances of central MUP isoforms in adulthood are stable . Differential transcription and silencing of central Mup genes among individuals is a major contributor to individual differences in patterns of urinary protein excretion . Previous discussions of chemical identity signatures have emphasized the importance of combinatorial information [3 , 24] , though the genetic mechanisms that give rise to combinatorial variation had not been elucidated . By dissecting the relative contributions of variation in coding sequence , transcription and translation we have identified an important role of transcription in increasing individuality ( Fig 5 ) . We identify mutations in the promoter regions of Mup genes associated with a lack of transcription ( Fig 4A ) , suggesting a mechanism through which regulatory changes lead to stable individual differences in patterns of urinary proteins . Chemical signatures of individual identity in other taxa , may similarly depend on regulatory variation to generate combinatorial semiochemical diversity . The individuality in scent signatures has been widely viewed as a cue that arises through neutral process [23] or balancing selection on traits for reasons unrelated to recognition such as immune function [24] . Mice use the identity information in urine markings in a range of mouse social and sexual behaviors including territorial interactions and mate choice [17 , 73 , 74] . Given the importance of identity information to mouse social behavior , it is reasonable to hypothesize that MUP diversity has been selected to facilitate distinctiveness [28] and has evolved to signal individual identity . Theoretical models of identity signal evolution show that when confusion is costly phenotypes should be favored when rare as this makes them more distinctive [25 , 26 , 28] , leading to frequency-dependent selection . The patterns of molecular variation presented here are consistent with a model where MUP diversity evolved to signal individual identity . While the present data suggest that social and sexual interactions have favored the evolution of increased identity information in house mice , the relative importance of different behaviors that make use of identity information in selecting for diversity remains unresolved . In addition to their role in individual recognition [3 , 5 , 17] , variation in central MUP isoforms has been shown to mediate assessment of heterozygosity [75] , avoidance of inbreeding [76] and cooperation with relatives [40] . Therefore , it is possible that assessment of heterozygosity or the avoidance of inbreeding via disassortative mating [77] could be the mechanisms that maintain MUP diversity . In theory , a preference for novel MUP patterns could drive the evolution of extreme polymorphism as has been suggested to contribute to male color polymorphism in guppies [78 , 79] . Preferences for novel signals and disassortative mating , however , are unlikely to explain the diversity of MUP phenotypes in light of the facts that ( i ) unlike female guppies , female mice actually prefer familiar rather than novel males [74]; ( ii ) in wild populations , females also produce individually distinctive MUP patterns in their urine [40 , 72]; and ( iii ) the identity information in MUP patterns is used outside of the context of intersexual selection by males in territory marking and defense [3 , 5] and females in shaping patterns of cooperative nesting [40] . Alternatively , the maintenance of the MUP signals used to assess heterozygosity and recognize close relatives may be maintained because of selection for individual recognition via MUPs [28] . Models of genetic kin recognition argue against the evolution of genetic identity signals because cooperation among individuals with shared phenotypes is expected to erode signal diversity [80–82] . In contrast , models suggest that individual recognition is capable of maintaining signal diversity as common phenotypes are expected to suffer costs of confusion [25 , 26 , 28] . Of course , these alternative behavioral mechanisms favoring diversity are not necessarily mutually exclusive . Modeling studies and behavioral experiments are needed to partition the relative importance of different social or sexual contexts in maintaining MUP individuality . Frequency-dependent selection may lead to balanced polymorphisms that maintain particular alleles for extended stretches of evolutionary time [83 , 84] . For example , correlated polymorphism in male coloration and mating tactics has been maintained for over 10 million years in multiple species of Uta lizards [85] . Alternatively , frequency-dependent selection coupled with high allele turnover when novel alleles are favored may result in accelerated divergence among alleles and lineages after speciation , as has been suggested for mating type recognition loci in fungi [86] and self-incompatibility alleles in plants [87] . The fact that central Mup gene sequences show signatures of purifying selection argues against a model of rapid allele turnover and diversification , though additional studies of Mup diversity in additional wild populations of Mus will be needed to clearly elucidate the evolutionary dynamics of the system . The dynamics may be complicated by selection on multiple gene sequences and regulatory elements as well as coding sequences . A significant portion of the regulatory variation seen among central Mup paralogs is likely controlled by mutations in the immediate upstream region of the gene ( Fig 4A ) . Despite many central Mup genes having apparently non-functional promoter sequences , they have not mutated into pseudogenes . This may be because they are expressed in other tissues at other times via a different set of regulatory sequences . They may also be maintained via ectopic gene conversion [88] , which has been suggested to be an important force in the evolution of the central Mup paralogs [41 , 55] . In contrast , a previous analysis of the central Mup genes using GENECONV found little evidence for gene conversion [89] , though as the authors of that analysis and others have pointed out , GENECONV can give misleading results when paralogous sequences are nearly identical , as is the case for the central Mup genes [89 , 90] . Comparing gene trees between subregions for incompatibilities , however , is a highly robust and powerful method for detecting gene conversion [90] . The fact that identical exon 1 sequences are coupled with divergent upstream sequences suggests a role for gene conversion reshuffling the arrangement of promoters and coding sequences in generating diversity . Thus , ectopic gene conversion may be a key mechanism for generating diverse expression profiles among Mup haplotypes . Silenced gene sequences may act to enlarge the mutational target size for distinctive amino acid variants that are subsequently expressed as a result of gene conversion between expressed and silenced exons or the rescue of a silenced gene by conversion of the promoter sequence [91] . The evidence for selection maintaining variation in cis-regulatory elements does not exclude the possibility that trans-acting elements are also encoded within the Mup locus . Breeding experiments designed to explicitly test the interaction of haplotypes from natural populations will provide additional insight into the biology of the Mup locus . The fact that the central Mup coding sequences as a whole are under purifying selection ( Fig 6A ) suggests that the overall protein sequence and structure are strongly conserved by selection . The restricted number of sites that show variation may arise due to constraints on the perception of urinary proteins . The individual identity signal in urinary protein blends is perceived by a subset of V2R receptors found in the vomeronasal organ with varying affinity to different isoforms [3] . Most of the non-synonymous variants in our sample are present among the paralogs in the mouse reference genome , though we did identify novel amino acid variants . Given the restricted geographic scope of our study , it is likely that broader sampling will reveal additional mutations as more wild mice are studied . Behavioral studies have suggested that mice are able to distinguish among many of the central MUPs encoded in the reference genome [3 , 5] and it is likely that the newly identified mutations produce perceptually distinctive proteins as well since the mutations are on the exterior of the protein ( S2 Fig ) . Constraints on protein form may also arise because of selection on the ligand-binding pocket [55 , 89] . In addition to being directly detected by the VNO , MUPs also bind volatile molecules with isoforms showing differences in their specific binding affinities [92 , 93] , which influence the volatiles held in scent marks [94] . Alterations to the coding sequence are only really beneficial if they bind to a distinctive set of V2R’s or influence volatile pheromone binding . The V2R’s belong to a large and dynamic gene family [95] but the subset related to detection of MUPs may similarly be under selection to be able to detect the diversity of pheromone blends in a population . Co-evolution between ligands and receptors has the potential to lead to rapid turnover [86 , 87] , though the purifying selection observed on the Mup genes suggests that in this case receptor-ligand coevolution may lead to greater phenotypic stasis . Whereas changes to coding sequences have the potential to produce either a non-functional protein or one that is not perceptually distinctive , any substantial changes to the expression levels of genes will produce a new and distinctive combination of MUPs . Differential transcription greatly increases individual identity in the population examined here and may be especially important in the evolution of combinatorial recognition signals such as the MUPs . As sequencing technology evolves , our ability to assemble full-length genes from complex multigene families from many individuals will improve , though we will still be faced with the challenge of interpreting patterns of variation . This problem is especially acute in tandemly-arrayed gene families such as the Mup locus where individual chromosomes differ in the number of gene copies present in the array [55 , 96] , complicating patterns of paralogous and allelic variation . To address how the diversity of the central Mup genes shapes individually distinctive phenotypes we collapsed allelic and paralogous variation within the central Mup genes ( Fig 2 ) . From the perspective of mice assessing phenotypic variation in the blend of MUPs excreted in the urine , whether novel protein isoforms are generated by allelic versus paralogous differences is irrelevant . The PD statistic ( Fig 2 ) provides a means for examining patterns of selection in such complex gene families . By comparing patterns of diversity in different parts of the genes , we provide evidence for frequency-dependent selection maintaining diversity in coding and regulatory sequences . The approach employed in this study holds promise for understanding the patterns of diversity in the Mup gene family in natural populations and may also be of use for other gene families where variation in the complements of proteins produced have fitness consequences , such as MHC . We provide molecular evidence for frequency-dependent selection maintaining MUP variation used for social recognition . As the variation among MUP isoforms is only known to function in scent communication [5 , 35 , 93 , 97] , our results suggests that social interactions are sufficient to maintain genetic variation in recognition phenotypes . Analyses of molecular variation indicate that rates of divergence among urinary protein sequences are constrained , likely by their interaction with V2R receptors and the ability to bind low molecular weight ligands . The combinatorial ratios of isoforms are known to be behaviorally relevant [3] , and accordingly our results highlight gene regulation as an important target of frequency-dependent selection in this system . Regulatory variation may be broadly important in maintaining individuality in combinatorial pheromone blends in other taxa as well .
We examined urinary protein variation in one F1 male descendant each from 18 unique crosses between wild house mice ( Mus musculus domesticus ) collected from the environs of Edmonton , Alberta , Canada ( S1 Table ) . The wild caught mice were transported to the Animal Care facility at the University of Arizona , where the F1 male offspring were produced . The animal care and breeding protocol was approved by the University of Arizona Institutional Animal Care and Use Committee . We collected multiple urine samples from the same individual over the course of many days in the lab ( see S2 Table for ages and sampling times ) . Upon being euthanized , mice were dissected , tissue was collected for subsequent extractions of DNA and RNA , and museum specimens were prepared . Specimens have been deposited at the Museum of Vertebrate Zoology at the University of California , Berkeley . We focused on male mice as MUP concentrations tend to be higher in males compared to females and the roles of MUPs in male territory marking are well understood [3 , 5 , 67 , 98] . Urine and tissues were collected from mice after they were sexually mature and had been given the opportunity to mate . Urine samples and liver RNA were collected weeks apart ( S2 Table ) . Previous research , however , has demonstrated that males maintain very stable MUP profiles in their urine throughout adulthood [5] . Indeed , our results indicate that the temporal separation of RNA and urine sample collection had minimal effect on the patterns of MUPs reported here . Urine was freshly collected from male mice by placing them in a clean cage; urine was pipetted once it was evacuated . Urine was briefly centrifuged to collect samples at the bottom of the tube and frozen at -80 C until subsequently shipped on dry ice to the University of Liverpool for analysis . Total protein concentration in mouse urine was measured using a Coomassie Plus protein assay kit ( Pierce , Rockford , USA ) , using bovine serum albumin as a standard . Urine samples were diluted in 0 . 1% formic acid to a concentration of 2 pmol/μl . All analyses were performed on a Synapt G1 Q-ToF mass spectrometer ( Waters , Manchester , U . K . ) , fitted with an ESI source and coupled to a Waters nanoAcquity UPLC . Samples were injected onto a MassPrep C4 desalting column before elution over a 10 minute stepwise acetonitrile gradient , acquiring spectra between m/z 300–2000 . Raw data was processed and transformed to a true mass scale using MaxENT1 maximum entropy software ( Waters Micromass ) . All data sets were processed at 1 Da/channel over a mass range of 18 , 300–19 , 000 Da , and a peak width of 0 . 6 was used to construct the damage model . The peak lists of the true mass spectra were saved as text files and imported into SpecAlign [99] for spectral alignment . Baseline subtraction was by a factor of 20 , an average spectrum was generated and spectra were aligned to this using the Combined ( Correlation and Peak matching ) Alignment method . We tested for individuality by examining phenotypic correlations of urine protein composition within and between individuals . In the urine samples from our study population there were 8 well-supported ESI/MS peaks that correspond to the masses ( in Da ) of central Mup urinary proteins: 18645 , 18650 , 18665 , 18683 , 18693 , 18708 , 18713 and 18724 . As our interest was in determining the consistency with which individuals reproduce the same blend of proteins , we used the ESI/MS peak heights associated with these protein masses to calculate the proportion of the total that was associated with each mass within a urine sample profile . We then calculated the correlation of the proportion of MUPs at each of the 8 masses between samples from the same and from different individuals and determined the Pearson correlation coefficients . Identity signals should be variable between individuals and similar within individuals so we tested for a difference in the correlation coefficient in the within versus between individual comparisons . Genomic DNA was extracted from spleen using a Gentra Puregene Tissue kit ( Qiagen ) . Traditionally , gene families have been studied by amplifying genes or exons using conserved primer sets followed by Sanger sequencing of clones [100 , 101] . This process is laborious and costly , especially for longer sequences requiring the design of internal primers and multiple rounds of Sanger sequencing . Additionally , cloning and Sanger sequencing has the potential to miss genetic variation without exhaustive sampling [100] . To reduce costs and to capture as much variation as possible , we devised a high-throughput sequencing approach for documenting variation among the Mup genes from multiple individuals derived from a wild population . Four pairs of primers were designed for conserved regions upstream and downstream of all the Mup genes , generating a ~5kb fragment . Assembly and alignment of the reads to mouse reference genome ( GRCm38 ) demonstrated that our PCR approach amplified most known members of the gene family ( only Mup3 and Mup21 , two divergent peripheral Mup genes , were not amplified ) ( Fig 1 ) . Furthermore , the same primers successfully amplified PCR products in samples from distantly related mouse species , such as Mus caroli . Together these two pieces of evidence suggest that our approach is likely to have captured the overwhelming majority if not all of the diversity in the central Mup genes . To guard against any potential effects of PCR biases in the amplification of paralogs , we amplified each of the 3 primer pairs ( S4 Table ) in triplicate ( for a total of 9 reactions ) before combining the samples and constructing sequencing libraries . PCR conditions were as follows: ( 1 ) initial denaturation at 98 C for 2 . 5 minutes , ( 2 ) 25 cycles at 98 C for 10 seconds , 58 C for 30 seconds , 72 C for 3 minutes , ( 3 ) a final elongation at 72 C for 7 minutes . We also used 1 primer designed to amplify peripheral MUP genes and amplified it using 1 round of PCR following the same conditions . We used a high fidelity long range Taq for PCR ( Platinum Taq , ThermoFisher ) . To construct libraries we sheared the uncleaned PCR products using a BioRuptor , targeting fragments in the 600–1000 bp range . Libraries were constructed using the sheared amplicons following a custom protocol for generating individually-barcoded genomic libraries for Illumina sequencing [102] . Libraries were sequenced at the University of California , Davis Genome Sequencing Center using one lane of MiSeq ( 300pb paired end mode ) . The MUPs excreted in urine are transcribed in the liver at high levels . At dissection , portions of the liver were stored in RNA-later kept at 4°C for one day and then frozen at -80°C until extracted . We extracted total RNA using the Qiagen RNeasy kit . As our goal was to sequence the liver mRNA with relatively long reads , we created double stranded cDNA libraries from RNA extracts rather than directly fragmenting RNA . Reverse transcription of full-length RNA has been shown to introduce biases into transcriptome libraries by failing to fully reverse transcribe long transcripts [103] . This is unlikely to influence our results because all the Mup transcripts are reasonably short and of similar length ( ~ 925 bp ) . While cDNA libraries can under-represent 3’ ends of long transcripts , all of the Mup transcripts are similar length and so should be similarly affected by any biases . As with the DNA amplicons , the cDNA samples were fragmented into 600–1000 bp range and then made into libraries using the Meyer-Kircher protocol [102] . Libraries were sequenced at the University of California , Davis Genome Sequencing Center using a MiSeq ( 300pb paired end mode ) . Aligning either DNA or RNA reads to reference sequences presents a significant problem in the attempts to study population level diversity of multigene families . The high levels of similarity among the central MUPs ( >98% coding sequence similarity ) mean that reads from one central MUP paralog readily map to all of the paralogs . Furthermore , the high levels of sequence similarity and gene conversion among central MUP paralogs preclude the unambiguous assembly of reads into distinct paralogs . The same issues are not faced by the more divergent peripheral MUPs , which have lower sequence similarity and map uniquely . To address alignment issues among the central Mup genes , we developed a novel approach for examining the DNA and RNA diversity among individuals in the population . Rather than aligning reads to the full complement of central Mup genes in the reference genome , we aligned reads to a single representative gene or transcript ( Mup11 ) from the mouse reference genome ( Fig 2 ) . For each site we were able to compare the percent of reads that have the reference versus non-reference base across individuals . From these data we calculated a pairwise difference statistic , PD , described in more detail below and in Fig 2 . We note that in the case of DNA , individuals vary in the number of copies of Mup genes [96] so that a given percentage of non-reference reads does not necessarily correspond to the same number of paralogs across individuals . For example , consider two individuals for which 20% of the reads are non-reference at a given site . If individual A has 25 central Mup paralogs across its two chromosomes and individual B has 20 , the number of paralogs with non-reference bases would be 5 and 4 respectively . Raw reads were pre-processed with Cutadapt [104] , FLASH [105] and Trimmomatic [106] to remove adapters , duplicate reads , contamination and to trim reads . Cleaned reads were aligned using default settings of Bowtie2 [107] . Commonly used variant calling programs assume a diploid state making them inappropriate for our alignments of multiple paralogs to a single reference paralog . We used the ‘count’ function in ANGSD to identify variants in our dataset [108] . The count function gives the number of reads reporting each base for a given site . The average read depth was >7000x for most individuals . Inspection of the distribution of read counts for each base showed very few sites with moderate read coverage for an individual ( e . g . ~100x where typical coverage depth was >7000x for an individual ) . Therefore we used a hard cut-off and masked all bases with less than 100 reads . This approach could lead to false negatives or false positives though these should not be any more common at some sites compared to others and therefore are unlikely to substantially influence our results or conclusions ( e . g . sequencing errors should occur evenly across non-synonymous and synonymous sites ) . The peripheral Mup genes are single copy genes in the mouse reference and are assumed to be so in this analysis . The following peripheral Mup paralogs were amplified by our PCR reactions: Mup4 , Mup5 , Mup6 , and Mup20 . Two genes , Mup3 and Mup21 , were not amplified by our PCR reaction , though sequences from Mup3 were available from liver transcriptomes . Therefore , we used standard variant calling methods in SAMtools [109] to generate a VCF file with the reads . We filtered putative SNPs based on quality , discarding any with a quality score of less than 100 . To understand the relationship between DNA sequence and RNA abundance in our samples we compared the reads covering exon 1 . We wrote a custom script to extract all of the reads that completely covered the exon ( 96–99 bp ) in each dataset type and grouped identical reads using CD-HIT [110] . The script outputs two files providing the sequence and read depth of each unique sequence respectively . True variants are expected to have high coverage while erroneous sequences should have low coverage when the sequencing error rate is low and coverage is high [111] . Following [111] , the list of putative exons for each individual was delineated by a sharp decline in read depth , with low coverage sequences assumed to be artifacts . We used the same procedure for determining putative exon 1 sequences in both the DNA and RNA datasets . All well-supported exon 1 sequences in RNA were independently identified as putative sequences in the DNA dataset , though some DNA sequences do not have corresponding sequences present in RNA ( Fig 3A ) . In principle , the MUP variants observed in the DNA but not in the RNA ( see Results and Fig 3 ) could be due to recombinant haplotypes artificially created during PCR , although several lines of evidence argue against this . First , previous studies have demonstrated that some MUP genes are not expressed [41] , consistent with our findings . Second , some of the MUP variants that are not expressed in some individuals are nonetheless expressed in other individuals in our dataset ( Fig 3A ) . Third , included among the variants that are uniquely found in the DNA are SNPs that are not found in the RNA; these cannot be a result of recombinant PCR . Finally we note , that the statistic PD does not take the phasing of haplotypes into account and so the potential for PCR recombinants would not affect the results of our population genetic analyses . To correct for differences in sequencing depth among individuals and between the DNA and RNA datasets , we calculated the proportional representation of each well-supported exon 1 sequence , after first filtering out the reads designated as artifacts . We considered the relationship between the proportional representation of a given sequence in the DNA and RNA datasets for all the exon 1 sequences in DNA as well as the reduced subset that show evidence of expression ( >1% of RNA reads within a given individual ) . Of the ten amino acid substitutions found among the expressed central Mup sequences ( S2 Fig ) , three cause only minor changes in molecular weight ( ~2 Da ) , preventing us from separating isoforms with and without these variants in ESI/MS spectra . Thus , we focused our comparison by collapsing the minor variation and considered the height of the following peaks: 18645 , 18650 , 18665 , 18683 , 18693 , 18708 , 18713 and 18724 Da ( S3 Fig ) . This approach also allowed us to focus on a 131bp region of the transcript spanning exons 2 and 3 . We assessed coverage depth of the 131bp region using a custom script that extracted all reads completely covering this region and grouped identical reads using CD-HIT [110] . From this output , we assigned the reads to an expected molecular weight and calculated the relative proportions of the weights . One amino acid variant outside of the 131bp region at the 3’ of the transcript ( Table 1 , S3 Fig , R161L ) also has a measurable effect on the molecular weight of proteins . To account for this variant , we subtracted the read depth associated with this variant from the percentage of the relevant reads from the 131 bp section . We compared the predicted proportion of molecular weights to the observed ratios of the predicted proteins at the relevant peaks . To examine the potential role of cis-regulatory mutations immediately upstream of the gene , we examined a ~325pb region including the first exon and approximately 200pb of upstream sequence ( S4 and S5 Figs; note that the sequence length varies due to an A-rich tract of variable length just upstream of the gene ) . We matched previously defined putative exons with promoter regions when there was a perfect alignment between exon sequences . The >300bp region is longer than a single read and is only fully covered when the forward and reverse reads of the same sequence overlapped creating a longer contig . Thus , coverage depth for reads fully-spanning this longer region was low compared to the shorter stretch used to identify exons or predict molecular weights , so methods to identify sequences based on high coverage were not applicable [111] . Therefore , we considered a sequence to be a putative promoter sequence if the full 300pb region was present more than once in the dataset . This resulted in the identification of the promoter sequences for most exon 1 sequences . We assigned promoter sequences as being expressed ( >5% ) , weakly expressed ( 1–5% ) or silenced ( <1% ) based on the proportional representation of the associated exon 1 in the RNA dataset . Many highly expressed exons are associated with multiple different promoters within a single individual , indicative of multiple gene copies . When expressed genes have a single promoter , one can unambiguously assign the promoter as being functional . Similarly , all promoter sequences associated with weakly or non-expressed sequences can be assigned to the lower expression category . By contrast , the presence of multiple promoter sequences makes it difficult to confidently ascribe function to any specific sequence . Therefore we did not consider groups of promoters associated with expressed genes . This left us with 20 promoters associated with exons with robust evidence of expression ( >5% ) and 112 promoters associated with weakly expressed or silenced exons ( <5% ) . The high bar for being considered expressed is unlikely to wrongly assign any silenced promoters as being expressed , though could potentially mis-categorize some expressed promoters as being non-expressed . We aligned all the promoter sequences and found greater levels of non-consensus bases among the silenced promoters compared to the expressed promoters ( S4 Fig ) . For each variant site , we used Fisher’s exact test to assess the association between the variant and expression phenotype . The position of variants relative to possible transcription factor binding sites in Fig 4 was determined using PROMO [112] with a cut off of 10% dissimilarity to the binding site motif allowed . As noted above , single exon 1 sequences are associated with multiple different promoter sequences within the same individual . To assess the overall relationship between promoter and coding sequences , we constructed neighbor-joining trees for each part of the sequence in MEGA 5 . 2 [113] and created a tanglegram showing the links between the promoters and exons using Dendroscope [114] . Since we were not able to align central Mup reads uniquely to reference genes , we could not use π or other widely used measures of sequence diversity to assess genetic variation within the sample . Instead , we calculated two metrics based on the logic of the site frequency spectrum and one based on π . We considered the distribution of variants in terms of the number of individuals in which the variant was detected as well as the overall percentage of reads in which it was detected . To calculate the overall percentage of reads , we summed the percentage of non-reference reads across individuals and divided by the number of individuals . Distributions of individual and percentage frequency spectra were compared among classes of sites using Wilcoxon tests . Pairwise sequence differences among individuals can provide information on the abundance and frequency of mutations . Importantly , differentiation among individuals is critical to allow for identification based on scent . A difference in the percentage of reads with a particular variant between two individuals is reflective of differences in the relative number of gene copies or alleles containing that variant between two individuals . We summed the total differences in the percentage of non-reference reads across individuals and termed this the pairwise difference index , PD , as it measures the extent of differences between two individuals ( Fig 2 ) . As the value of PD is determined for each pair of individuals , we chose to represent it by its distribution across all pairwise comparisons in our dataset ( N = 153 ) and compared differences in the distributions using Wilcoxon tests . We use the notation PD’ to denote values that compare the total difference between two individuals uncorrected for the number of sites examined and PD to denote site-corrected comparisons . We used individual-based forward time simulations to provide additional tests of selection for our dataset . Simulations were written in C++ and the source code is available in S1 File . Briefly , we simulate a Wright-Fisher population of diploid individuals wherein each chromosome has a cassette of 15 MUP proteins as in the reference genome . All rate parameters are based on expected values for Mus scaled by 100 . We then allow gene conversion between any two MUP paralogs at a rate of 1e-4 with a conversion tract length drawn from a geometric distribution of mean 50 . Recombination occurs between homologous MUP proteins at a rate of 5e-6/cassette/individual/ generation . These conditions fix the cassette size at 15 . Although the cassette length undoubtedly varies in natural populations [96] , and this may be an important source of functional variation for diversifying selection , there are few models of cassette size mutation precluding us from analyzing this aspect of functional evolution of MUP proteins . We then allowed mutations at 15 total non-synonymous sites and 102 synonymous substitutions at a rate of 3e-6/site/generation . All other non-synonymous sites are assumed to be deleterious and are therefore not expected to experience diversifying selection , consistent with sequence similarity of MUP proteins . Negative frequency dependent selection is applied as follows: every generation the mean frequency per individual of each non-synonymous mutation is computed . We then compute the total difference in frequency between each individual’s genotype and the mean genotype across all non-synonymous sites . The fitness of the maximally different individual is equal to one , and an individual of exactly average allele frequencies is fitness 1 –s . The fitness of all other individuals is based on their total allele frequency differences relative to the average frequencies and distributed linearly between 1 and 1—s . We conducted 10 , 000 simulations wherein we drew s from a uniform distribution between 0 and 0 . 1 , and the effective population size from a uniform distribution between 100 and 1 , 100 . These prior bounds were selected after 1 , 000 simulations with much wider priors suggested the maximum a posteriori estimate would be within each interval . Each simulation was run for 10N generations . We then used ABCreg [115] to regress the following four summary statistics onto the data: the number of non-synonymous and synonymous segregating sites in a sample of 18 individuals , and the mean frequency of synonymous and non-synonymous substitutions . We used the tangent transformation in ABCreg , and defined the tolerance at 0 . 02 to obtain the posterior distribution of selective coefficients . We estimated the maximum a posteriori estimate using the locfit package in R [116] . In an effort to be conservative in our inference of selection operating on non-synonymous polymorphisms in Mus , we also performed an additional 10 , 000 neutral simulations . We then combined these neutral simulations with the 10 , 000 simulations described above to produce a set of results based on a highly neutrality-biased prior distribution . Based on this very conservative prior distribution , and using the same regression approach as above , the resulting credible interval still does not include s = 0 . | Individual recognition via scent is critical for many aspects of behavior including parental care , competition , cooperation and mate choice . While animal scents can differ in a huge number of dimensions , recent work has shown that only some specialized semiochemicals in scent marks are behaviorally relevant for individual recognition . How is individuality in specialized semiochemical blends produced and maintained in populations ? At the extremes , individuality may depend on either a plethora of semiochemical isoforms or on combinatorial variation in a small number of shared isoforms across individuals . Analyzing the major urinary protein ( MUP ) pheromone blends of a wild population of house mice , we find evidence in favor of a combinatorial diversity model for the production and maintenance of individuality . Balancing selection maintains MUP proteins at moderate frequencies in the population , though interactions with the pheromone receptors appear to limit the extent of pheromone diversity in the system . By contrast , differential transcription of proteins greatly increases individuality in pheromone blends with balancing selection maintaining diversity in promoter regions associated with gene expression patterns . Selection maintaining combinatorial diversity in a limited set of behaviorally important semiochemicals may be a widespread mechanism generating and maintaining individuality in scent across taxa . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] | [
"sequencing",
"techniques",
"medicine",
"and",
"health",
"sciences",
"body",
"fluids",
"population",
"genetics",
"dna",
"transcription",
"urine",
"sequence",
"motif",
"analysis",
"molecular",
"biology",
"techniques",
"population",
"biology",
"mammalian",
"genomics",
"research",
"and",
"analysis",
"methods",
"sequence",
"analysis",
"sequence",
"alignment",
"gene",
"expression",
"molecular",
"biology",
"animal",
"genomics",
"biochemistry",
"anatomy",
"physiology",
"genetics",
"biology",
"and",
"life",
"sciences",
"pheromones",
"genomics",
"evolutionary",
"biology"
] | 2016 | Selection on Coding and Regulatory Variation Maintains Individuality in Major Urinary Protein Scent Marks in Wild Mice |
The spread of tick-borne pathogens represents an important threat to human and animal health in many parts of Eurasia . Here , we analysed a 9-year time series of Ixodes ricinus ticks feeding on Apodemus flavicollis mice ( main reservoir-competent host for tick-borne encephalitis , TBE ) sampled in Trentino ( Northern Italy ) . The tail of the distribution of the number of ticks per host was fitted by three theoretical distributions: Negative Binomial ( NB ) , Poisson-LogNormal ( PoiLN ) , and Power-Law ( PL ) . The fit with theoretical distributions indicated that the tail of the tick infestation pattern on mice is better described by the PL distribution . Moreover , we found that the tail of the distribution significantly changes with seasonal variations in host abundance . In order to investigate the effect of different tails of tick distribution on the invasion of a non-systemically transmitted pathogen , we simulated the transmission of a TBE-like virus between susceptible and infective ticks using a stochastic model . Model simulations indicated different outcomes of disease spreading when considering different distribution laws of ticks among hosts . Specifically , we found that the epidemic threshold and the prevalence equilibria obtained in epidemiological simulations with PL distribution are a good approximation of those observed in simulations feed by the empirical distribution . Moreover , we also found that the epidemic threshold for disease invasion was lower when considering the seasonal variation of tick aggregation .
Several ecological studies have shown that the distribution of ticks on their hosts is often highly aggregated , with a large number of hosts harbouring few parasites and a small number harbouring a large number of them ( [1]–[5]; other interesting references could be found in [6] ) . In addition , the distribution of tick development stages is coincident , rather than independent [7] . Specifically , those hosts feeding larval tick stages were simultaneously feeding the greatest number of nymphs . As a result , about of all hosts feed of both larvae and nymphs and the number of larvae feeding alongside nymphs is twice as many as it would be if the distributions were independent [7] , [8] . The aggregation of parasites on hosts bears important implications for vector-borne disease dynamics , since the small fraction of hosts supporting the bulk of the vector population is also responsible for the majority of the pathogen transmission [9] . The transmission of tick-borne diseases is characterised by an intricate set of ecological and epidemiological relationships between pathogen , tick vector , vertebrate hosts and humans that largely determine their temporal and spatial dynamics [10] . Tick-borne disease dynamics feature several complexities , due to the presence of a number of heterogeneities in the system coupled with non-linear phenomena operating in the transmission processes between ticks , host and pathogen [11] . The transmission of pathogens from one tick to another , a pre-requisite for the establishment of cycles of infection , may occur via three different pathways depending on the pathogen ( see [12] for a comprehensive review ) . First , adult female ticks may transmit the pathogen to eggs trans-ovarially . Second , ticks may infect a host during their blood meal , leading to a systemic infection in the host; ticks might then acquire the infection by feeding on an infected host , maintaining the infection trans-stadially . Third , ticks may become infected by co-feeding with infected ticks on the same host . Co-feeding transmission is also called non-systemic as it does not require the host to have a systemic infection , since pathogens are transmitted from one tick to another as they feed in close proximity . Vertebrate hosts may vary in their competency to support systemic and co-feeding transmission [13] . Tick-borne pathogens differ also for the mechanisms which they use to persist in nature . For instance , Rickettsia spp . , the pathogen agents causing Rocky Mountain Spotted Fever , are maintained by systemic and trans-ovarial transmission in Dermacentor variabili and andersoni [14] while it has been observed that Borrelia Burgdoferi s . l . spirochaetes persist in nature by taking advantage of all three routes of transmission in I . ricinus , [13] , [15] . In the case of the tick-borne encephalitis virus ( TBEv ) , which is an increasing public health concern in Europe [16]–[18] , trans-ovarial transmission seems to be relatively rare and its contribution is generally thought to be negligible [19] . On the other hand , both systemic and non-systemic transmission can take place on reservoir-competent rodent hosts . However , due to the very short duration of the TBEv infection in rodents , [20] , the systemic route would only allow infection of a very limited number of ticks . Indeed , non-systemic transmission through co-feeding ticks is a more efficient transmission route for TBE [8] , [20] . Different studies have shown that TBEv would not become established in competent hosts , such as rodents , without the amplification of the overall transmission efficiency provided by co-feeding transmission ( see for instance [20]–[23] ) . The aggregation pattern of ticks on hosts therefore plays a more important role in the transmission of TBEv than in other tick-borne pathogens , such as Borrelia burgdorferi sensu lato and Anaplasma phagocytophilum , where other efficient routes of transmission have been observed . Tick aggregation on hosts and correlation of tick stages facilitate co-feeding transmission and thus significantly increase the basic reproductive number , , of the pathogen , with direct implications for its persistence [23] , [24] . Using different levels of aggregation ( from independent to coincident aggregated distribution ) , Harrison and collaborators [23] showed that values of increase with progressive levels of aggregation , making it more likely for tick-borne pathogens to become established and persist . In addition , the authors of the cited works evinced that when ticks followed a coincident aggregated distribution , the increase of was greater than in the case of independent aggregated distributions . The degree of aggregation of ticks can be measured in a number of ways . Since the appearance of influential works by Randolph [25] and Shaw et al . ( [6] and [26] ) the negative binomial ( NB ) distribution has been extensively used to describe tick aggregation on hosts ( see e . g . [23] , [27] , [28] ) . Alternatively , other works suggested that different distributions characterised by larger tails than NB ( i . e . , predicting more rodents with very large tick burden than expected with NB ) , can be effective in describing tick aggregations . Specifically , a Poisson-LogNormal ( PoiLN ) mixed model has been successfully used to describe tick distribution on red grouse chicks [29] , while Bisanzio and collaborators [30] showed the first evidence that the distribution heterogeneity of ticks on hosts seemed to be better described by a power-law ( PL ) than a negative binomial distribution . A suitable description of the distribution tail might have important consequences on the dynamics of the pathogen spreading process . Modelling the spread of vector-borne diseases through bipartite networks [30] showed that the extreme aggregation of ticks on hosts has dramatic consequences on the behaviour of the epidemic threshold . In the current study we used an extensive data set of Ixodes ricinus ticks feeding on mice ( a total of 4722 parasitised hosts collected in 9 years ) to detect the best fit for the distribution of tick burden on mice by testing the performance of NB and PoiLN versus PL distribution , with particular interest in the shape of the distribution tail which is crucial to suitably describe the fraction of co-feeding ticks necessary for TBEv transmission . Then , we used a stochastic model to simulate the effect of fitting different tick distributions on the infection dynamics of a tick-borne pathogen . Specifically , we investigated the spread of a non-systemically transmitted pathogen ( e . g . TBEv ) by modelling the pathogen transmission between susceptible and infective ticks , considering only co-feeding transmission and distributing ticks on mice under the hypotheses of NB , PoiLN , and PL distributions . Finally , we investigated the seasonal variations in the pattern of tick burden distribution on mice and its implication on TBE-like infection dynamics .
All animal handling procedures and ethical issues were approved by the Provincial Wildlife Management Committee ( renewed authorisation n . 595 issued on 04 . 05 . 2011 ) Rodent tick burden data was collected by trapping mice using capture-marking-recapture techniques during 2000–2008 . The study area was a mixed broadleaf woodland [7] , [21] , located in Valle dei Laghi within the Autonomous Province of Trento , in the north-eastern Italian Alps ( grid reference 1652050E 5093750N , altitude 750–800 m a . s . l . ) . In the year 2000 , mice were monitored in nine selected areas through placement of 8×8 trapping grids with a 15-m inter-trap interval . In 2001 and 2002 the number of trapping grids was reduced to eight , while from 2003 onward their number was further reduced to four . In summary , the trapping effort consisted of twice-daily trap sessions with at least one capture , resulting in a total number of Apodemus flavicollis captured with at least one tick attached . For each captured rodent the number and life stage of feeding ticks was carefully assessed and registered , without removal [7] , [21] . A total number of ticks were counted of which were larvae , were nymphs , and were adults . The number of ticks [nymphs] per rodent was between 1 [0] and 111 [15] with a median number of ticks per rodent equals to 8 . Detailed data , on a yearly scale , are reported in Table 1 , while the fraction of nymphs observed in different year and grids is reported in Table 2 . In Figure 1 the number of captured Apodemus flavicollis per trapping session is shown for the whole nine year period and for different grids ( from A to I ) . In order to explore the impact of different parasite aggregation distributions on the spread of a TBEv-like pathogen where the main transmission route is through co-feeding , we performed extensive numerical simulations informed by the data about tick aggregation on mice . In this setting , tick larvae were not infective ( transovaric transmission has been indicated as negligible [38] ) , adults only rarely feed on mice ( on our data set adults ticks are about of the total number of ticks feeding on mice ) , and the only transmission link that we considered was the co-feeding between infective nymphs and larvae . Therefore , the only actors in our model were nymphs and larvae feeding on hosts . Moreover , Rosà and collaborators suggested in a recent work devoted to the same geographical area [21] that the larvae that feed in one year generally quest and feed as nymphs in the following year . Therefore , by adapting the Susceptible-Infected-Susceptible ( SIS ) model [39] to our purpose we assumed that nymphs are categorised as infective or not , that feeding larvae are susceptible and that some of them could eventually be infected by co-feeding with infective nymphs before moulting ( thus becoming infective nymphs at time t+1 ) . At each iteration t , with t being a discrete number between and and year , we assigned a number of ticks to each of the mice by drawing a sample from the considered distribution q . Then , on each mouse we said that of ticks feeding on it , were nymphs and the other larvae ( with ) . These nymphs were larvae in the previous year and were possibly infected . Then , defining as the prevalence among larvae after feeding at time , we assumed that the prevalence at time t among nymphs was . Thus , the number of infective nymphs on a mouse that at time t was parasitised by ticks was . Then , on each of the mice the co-feeding transmission between larvae and infective nymphs could occur with probability and we updated accordingly to the fraction of larvae infected ( i . e . the fraction of infective nymphs at next time step ) . The following meta-code summarises the epidemiological dynamic 1 . for t between and : ( a ) for each mouse i , with i between 1 and ( b ) is updated as the fraction of larvae infected ( c ) if is equal to zero we stop the loop It is worth stressing that in the previous meta-code we did not consider ticks recovering from the infection , since we assumed that a feeding infective nymph at time t will exit the infectious dynamics by moulting to the adult stage or dying . We also modified the previous dynamics to deal with different distributions in tick aggregation as a function of seasonality . At each year t , we classified mice as observed during the mice peak activity ( = mice , with ) and observed out of the peak ( ) . Therefore , we assigned the number of ticks feeding on mice according to the respectively aggregated distributions qIN and qOUT . Moreover , since the larvae obtaining a blood meal at year t will be nymphs at year without any other involvement in the epidemic spreading at year t , [21] , these modifications to the meta-code are sufficient to suitably describe the seasonal variation in the epidemic process . More explicitly , the epidemic dynamic in the presence of seasonality in tick aggregation may be described by the following meta-code: 1 . for t between and :
The probability distribution of tick burden on mice was skewed and showed a heavy tail . The best fit of the NB distribution was obtained on the largest available subsets of data , i . e . with , see left panel of Figure 3 . In this setting , the MLE method estimated ( confidence intervals ( CI ) ) and ( 95%CI = ) . However , the GOF of the NB distribution was very low for any value of , see central panel of Figure 3 , thus giving evidence for rejecting the hypothesis of the NB functional form . Similarly , the best fit of PoiLN distribution was achieved on the largest subsets of data , ( , see left panel of Figure 3 ) . In this case the estimated parameters were ( CI = ) and ( CI = ) . The GOF of the PoiLN , central panel of Figure 3 , suggested that PoiLN was acceptable only for . However , for , the KS statistic displayed values that were too large to consider the PoiLN distribution appropriate for describing real data . On the other hand , by fitting the tail of the distribution to a PL distribution , we found that the best fit was obtained for ( with a standard deviation of 5 . 83 ) , see left panel of Figure 3 . This value is matched with an estimated scaling parameter ( with standard deviation = 0 . 41 ) . The GOF test ( p-value larger than 0 . 1 ) suggested that the optimum PL fit on the tail of the distribution should not be ruled out , and that the result holds for every PL fit with see center panel of Figure 3 . Finally , the LLR test highlighted that the PL fitting is to be preferred ( ) to the NB in describing the tail of the distribution for a large range of lower bounds , , see right panel of Figure 3 . Similarly , the PL is to be preferred to the PoiLN for . Moreover , it is worth to stress that for values above ( 55 ) the sign of the LLR test still indicates the PL fit as the preferred one compared to the NB ( and PoiLN ) , although the indication loses statistical significance due to the scarcity of available data . In Figure 4 we show the complementary cumulative probability distribution of the best fits resulting from for NB and PoiLN distributions and for PL distribution against field data of the number of ticks per mouse . From this plot we noticed that above a certain number of ticks per mouse NB [PoiLN] under-estimates [over-estimates] the tail of the distribution ( indeed both fits were statistically evaluated as very poor ) . At the same time , in agreement with statistical results summarised in Figure 3 , we noticed that the PL fit in Figure 4 more appropriately describes the right tail of the data distribution . The number of mice captured in different years and grids showed strong seasonal patterns as reported in Figure 1 . For each grid and each year we defined two separate periods depending on the mice abundance as defined in section “Data Analysis” and sketched in Figure 2 . Imposing a threshold , for each year and grid we identified a time window of high mice abundance . With we found significant evidence that the distribution of ticks on mice within the abundance peak was different from that observed outside . Indeed , the fraction of the KS measures calculated on the synthetic samples lower than the real-data KS statistic was almost , thus indicating very low confidence in obtaining the same measurement by chance . The same statistical evidence was also obtained by using different time window thresholds ( such as , and 0 . 6 ) . On the data sets classified as inside ( IN ) and outside ( OUT ) the time window of mice abundance peak , we fitted for different time-window lengths ( ) the parameters and for NB distribution ( Figure 5 , left panels ) and and for PL distribution ( Figure 5 , right panels ) . We observed a larger PL scaling parameter inside the mice abundance peak than outside ( two-sample t-test output: for t-statistic = , df , ) indicating a larger heterogeneity in tick burden outside the abundance peak time . Moreover the GOF test indicated a rejection of the NB fit in both sets ( IN and OUT ) with . On the other hand , the GOF test with showed that the PL model cannot be ruled out in both sets ( p-value>0 . 1 ) and the LLR test indicated that the PL fitting outperforms the NB model ( p-value<0 . 05 ) in the estimates both inside and outside the peak time window . The distribution of larvae and nymphs on mice are coincident rather than independent , and indeed the same most infested hosts feed both of the nymphs and of the larvae . Moreover , Spearman's correlation coefficient measured on the number of larvae and nymphs on mice was positive ( 0 . 24 ) and the probability that this coefficient was detected by chance was very low ( the empirical value was the largest if compared to those evaluated in reshuffled samples ) . In addition , the mean number of larvae co-feeding with a nymph is about which is almost double the value that would be seen if the distributions were independent ( mean equal to 12 ) . To start , we simulated the non-systemic disease spreading of a TBE-like pathogen with a fraction of nymphs among ticks equals to 2% , close to the one observed in our real data ( cfr . Table 2 ) , 5% , and 10% , as in literature [8] , [40] . We consider the empirical distribution observed on the entire data set . We fixed the number of hosts to which , together with the considered distribution , resulted in a number of vectors pairs equal to . In our simulations , we explored the effects of β , the infection probability , on the observed prevalence at the final time step , , with . ( We observed that was larger enough to allow the prevalence to converge toward an endemic pseudo-equilibrium or the disease-free equilibrium ) . For each we allowed simulations to run starting from an initial prevalence of . In Figure 6 we plotted the prevalences ( median value , interquartile intervals and the CI ) observed at equilibrium as a function of the transmission probabilities , β . Results showed that the larger the fraction of nymphs among ticks feeding on mice , the larger the probability of pathogen invasion and the infection prevalence . Then , we explored the effects of different tick burden distributions on the spread of infection . To this end we considered four distributions: PL , NB , PoiLN , and the empirical distribution on the entire data set ( aggregated on capture sessions and grids ) . For synthetic distributions we considered the actual observed distribution below the estimated , while we used the best fit of synthetic distributions to describe values greater than . Again , we fixed the number of hosts to . It is worth stressing that in the synthetic samples generated from these distributions we observed some features similar to those observed in real-data . For instance , the number of nymphs was positively associated with that of larvae and more particularly a nymph co-fed with a mean number of larvae similar to that observed in reality ( for PL the mean number was 23 , for NB 20 , and for PoiLN 27 ) . Results , plotted in Figure 7 for and in Text S2 for and , corroborated the hypothesis that the transmission probability needed for the pathogen to become endemic is driven by the shape of the tail of the distributions . In particular , we noticed that for the PoiLN distribution ( the one with larger fitted tail ) the epidemic threshold is the lowest , while for the NB distribution ( the one with smaller fitted tail ) the infection probability needed for invasion is the highest . Not surprisingly , the PL , which has the best performances in fitting the tail of the empirical distribution , is the one for which the prevalences at equilibria better resemble those observed in simulations using the empirical distribution . We also performed some sensitivity analysis on parameter distributions , further highlighting that the larger the tail of the distribution , the lower the epidemic threshold ( see Text S1 ) . In addition , sensitivity analysis on the fraction of nymphs ( f ) showed that does not qualitatively influence the epidemic behaviour ( see Text S2 ) . Furthermore , we investigated the effect of differences in the distribution of the tick burden as a function of the abundance of mice on the spreading of a non-systemic infectious disease . To this end , we fixed , as measured in the dataset , and as qIN we considered a PL with exponent as estimated with . In a similar way , we assumed as qOUT a PL distribution with exponent . For both qIN and qOUT we further set . Results are summarised in Figure 8 , from which it could be inferred that the epidemic outcome was strongly influenced by the different distributions of feeding ticks according to mice abundance . We consistently observed that the transmission probability needed for the pathogen to effectively spread was smaller when the time windows identified by mice abundance are considered .
Tick aggregation on hosts is the result of several complex interactions of biotic and abiotic factors , such as host exposure and susceptibility to ticks , ticks' phenology and host behaviour , environmental factors , availability of resources , and others [27] , [41] . Historically , the NB distribution has been preferred to the Poisson distribution to describe parasite heterogeneity across hosts because it suitably reproduces overdispersed observations . It has also been widely used in empirical [6] , [25] , [26] , [28] and theoretical studies [23] , [24] , [42] . However , fat tailed distributions other than the NB one can also adequately reproduce tick aggregation , as shown by Elston et al . [29] and Bisanzio and collaborators [30] . Through the use of an extensive data set of feeding Ixodes ricinus ticks on mice , we showed that a PL distribution is better able to describe the right tail of the tick distribution on hosts than a NB or a PoiLN distribution ( see Figure 3 and 4 ) . This finding may have relevant epidemiological consequences , since it is well documented that the heterogeneity of contact distributions among individuals has large impacts on pathogen spread and persistence [43]–[49] . In fact , it has been demonstrated [50] that the minimum transmission probability for a pathogen to spread on a network , the so-called epidemic threshold , is driven by the first and the second moment of this distribution . In particular , Pastor-Satorras et al . [50] demonstrated that the larger the heterogeneity , the lower the epidemic threshold for the pathogen to spread , with an interesting behaviour in infinite size network showing a zero epidemic threshold [46] . Thus , the epidemiological inferences on the spread of a pathogen are highly influenced by the characterisation of the connectivity distribution and in particular by the distribution tail ( i . e . the heterogeneity ) . Our results corroborate those findings and generalise them in a different framework and for more complex transmission routes , i . e . a vector-host network for non-systemically transmitted diseases . In particular , we found that the tail of the distribution of the number of ticks per rodent highly influences pathogen spreading ( see Figure 7 and Text S1 ) . Furthermore , it is worth remarking that although the tail of the distribution as defined here represents about of the entire data set , our simulation findings suggest that this small part of the distribution is crucial for pathogen invasion . We also confirm that the probability of pathogen invasion and the infection prevalence are strongly influenced by the fraction of nymphs on the total feeding ticks on mice ( Figure 6 and Text S2 ) . The co-occurrence of larvae and nymphs on competent hosts is in fact essential for the horizontal transmission of non-systemic transmitted tick-borne pathogens , such as TBE , and it has been documented , both empirically and theoretically , that it could be a key factor in creating TBE hotspots , [51] , [52] . Our conclusions confirm previous findings showing that the distribution of ticks on rodents may significantly affect the spread of infections [27] , [30] , [53] , especially for non-viraemic transmitted diseases such as TBE [7] , [23] , [24] . Under the hypothesis of a NB distribution of ticks across hosts , both Rosà et al . [24] and Harrison and collaborators [23] showed that highly coincident and aggregated distributions favour the establishment of TBEv . However , highly heterogeneous degree distributions do not necessary imply a higher spread of disease . Indeed , Piccardi et al . [54] showed that scale-free networks can be much less efficient than homogeneous networks in favouring the disease spread in the case of a nonlinear force of infection . The correct description of tick aggregation on hosts could dramatically affect disease control strategies: for instance , Perkins [7] emphasised that an optimised control effort targeted on highly parasitised mice , also identified as sexually mature males of high body mass , could significantly lower the transmission potential . On the other hand , Brunner and colleagues [27] observed that the identification of individuals which fed a disproportionate number of ticks ( and that can therefore act as superspreaders ) can be challenging , since simple covariates such as sex , age or mass do not entirely explain the differences in parasite burden . In order to fully understand the different tick attachment behaviours on hosts , we identified different time windows related to rodent seasonal dynamics . Using this approach we found that the distribution of ticks on mice may vary across the season , with higher aggregation heterogeneity in periods of low rodent abundance and lower aggregation heterogeneity during the peak of host abundance ( see Figure 5 ) . We also showed that seasonal aggregation patterns , characterised by larger tails in time periods of low host abundance , enhance the spread of non-viraemic transmitted diseases ( see Figure 8 ) . Shaw and collaborators [26] observed significant variations in the degree of aggregation between host subsets – stratified by sex , age , space or time of sampling – in several host-parasite systems . In agreement with our results ( lower aggregation in period of high mice abundances as shown by estimated exponents of PL ) , they found that aggregation in copepod ( Lepeophtheirus pectoralis ) infesting plaice ( Pleuronectus platessa ) decreases during summer months . They mainly ascribed the observed variation to significant differences in mean parasite burden among months . On the other hand , we did not find significant differences in tick burden inside and outside the window of high rodent abundance . Specifically , in the case of , the average number of ticks per host were inside the window of high rodent abundance and outside and the differences between inside and outside are not statistically significant ( permutation tests , p>0 . 05 ) . However , the second moment of the number of ticks per host drastically changed between high and low abundance periods , driving the difference in the aggregation distributions observed in the two time windows . Seasonal variations in resource availability and host abundance can have a significant effect on the space used by mice . Males and females tend to respond to these changes in different ways , since space use for females is driven largely by food availability , whereas the distribution of males is related primarily to mating opportunities . Yellow-necked mouse ( A . flavicollis ) females exhibited reduced spatial exclusivity and larger home ranges during lower food availability while males varied their spatial distribution accordingly by also expanding their home ranges [55] . An inverse relationship between population density and home range sizes has also been observed in wood mice ( Apodemus sylvaticus ) [56] . Consequently , in periods of low rodent abundance more mobile rodents , especially males , are more likely to hit a patch of larval ticks . As a result , these individuals would harbour a large amount of ticks and increase the aggregation of tick distribution among the rodent population . On the other hand , tick density is usually lower in periods of low rodent abundance , and the average tick burden would decrease for the rest of the population , especially females , balancing the overall tick burden . On the contrary , during times of high abundance mice move less and ticks would be distributed more evenly among the rodent population resulting in the observation of a lower aggregation in tick distribution during the peak of rodent abundance . Our primary goal was to help understand the role of tick aggregation across mice on the spread of non-viraemic transmitted diseases through a simple and general transmission model . Other works – such as [24] , [42] , [52] , [57] – described in very fine detail the transmission of vector-borne diseases , introducing different transmission routes , tick stages and alternative hosts in the epidemic model . For instance , Norman and colleagues [57] demonstrated through an epidemiological model that non-viraemic transmission could have non-negligible effects on the persistence of a disease like the Louping ill . Here , considering the non-systemic transmission only , we explored the effect of using different theoretical functional forms to describe the tick burden on hosts . By estimating parameters of the burden distributions on a very detailed data set , we defined a simple and transparent transmission model that explicitly takes into account the real contact pattern of vectors and hosts in the description of a non-systematically transmitted vector-borne disease . In this way we were able to emphasise that , while the NB and PoiLN models can sufficiently fit the whole real distribution , the PL model represents a better fit for the distribution tail . Furthermore , the vector perspective approach used in our model gives better insights into the dynamics of non-systemic transmitted pathogens respect to host perspective models that were more commonly and widely used in this context [24] , [42] , [52] , [57] . In addition , epidemiological simulations parameterised by the fitted tick burden distributions highlighted the epidemiological consequences of describing tick aggregation on hosts trough distributions with different tails , showing that the shape of the tail distribution has a non-negligible influence on pathogen persistence . Future works will be devoted to extend the present findings to more complex transmission dynamics ( e . g . including viraemic or transovaric transmission ) , in order to assess the effect of a PL decay of the distribution for a wider range of vector-borne diseases . | Our work analyses a 9-year time series of tick co-feeding patterns on Yellow-necked mice . Our data shows a strong heterogeneity , where most mice are parasitised by a small number of ticks while few host a much larger number . We describe the number of ticks per host by the commonly used Negative Binomial model , by the Poisson-LogNormal model , and we propose the Power Law model as an alternative . In our data , the last model seems to better describe the strong heterogeneity . In order to understand the epidemiological consequences , we use a computational model to reproduce a peculiar way of transmission , observed in some cases in nature , where uninfected ticks acquire an infection by feeding on a host where infected ticks are present , without any remarkable epidemiological involvement of the host itself . In particular , we are interested in determining the conditions leading to pathogen spread . We observe that the effective transmission of this infection in nature is highly dependent on the capability of the implemented model to describe the tick burden . In addition , we also consider seasonal changes in tick aggregation on mice , showing its influence on the spread of the infection . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] | [
"ecology",
"and",
"environmental",
"sciences",
"medicine",
"and",
"health",
"sciences",
"biological",
"data",
"management",
"infectious",
"disease",
"epidemiology",
"population",
"dynamics",
"applied",
"mathematics",
"epidemiological",
"methods",
"and",
"statistics",
"plant",
"science",
"systems",
"science",
"mathematics",
"probability",
"distribution",
"statistics",
"(mathematics)",
"population",
"modeling",
"plant",
"pathology",
"theoretical",
"ecology",
"population",
"biology",
"distribution",
"curves",
"biostatistics",
"infectious",
"disease",
"control",
"veterinary",
"science",
"statistical",
"distributions",
"complex",
"systems",
"infectious",
"diseases",
"computer",
"and",
"information",
"sciences",
"veterinary",
"diseases",
"zoonoses",
"veterinary",
"epidemiology",
"metapopulation",
"dynamics",
"theoretical",
"biology",
"ecosystems",
"probability",
"density",
"epidemiology",
"infectious",
"disease",
"modeling",
"population",
"ecology",
"probability",
"theory",
"behavioral",
"ecology",
"ecosystem",
"modeling",
"ecology",
"disease",
"dynamics",
"veterinary",
"informatics",
"biology",
"and",
"life",
"sciences",
"physical",
"sciences",
"computational",
"biology",
"statistical",
"methods"
] | 2014 | Pattern of Tick Aggregation on Mice: Larger Than Expected Distribution Tail Enhances the Spread of Tick-Borne Pathogens |
Supported by recent computational studies , there is increasing evidence that a wide range of neuronal responses can be understood as an emergent property of nonnegative sparse coding ( NSC ) , an efficient population coding scheme based on dimensionality reduction and sparsity constraints . We review evidence that NSC might be employed by sensory areas to efficiently encode external stimulus spaces , by some associative areas to conjunctively represent multiple behaviorally relevant variables , and possibly by the basal ganglia to coordinate movement . In addition , NSC might provide a useful theoretical framework under which to understand the often complex and nonintuitive response properties of neurons in other brain areas . Although NSC might not apply to all brain areas ( for example , motor or executive function areas ) the success of NSC-based models , especially in sensory areas , warrants further investigation for neural correlates in other regions .
Brains face the fundamental challenge of extracting relevant information from high-dimensional external stimuli in order to form the neural basis that can guide an organism's behavior and its interaction with the world . To support complex patterns of behavior , populations of interconnected neurons must implement a rich repertoire of linear and nonlinear operations on their synaptic inputs that take into account context , experience , and anatomical constraints [1] . For example , anatomical bottlenecks often force the information stored in a large number of neurons to be compressed into an orders-of-magnitude-smaller population of downstream neurons [2–4] , such as storing information from 100 million photoreceptors in 1 million optic nerve fibers or resulting in a 10–10 , 000-fold convergence from cortex to the basal ganglia [3] . One potential approach to addressing this challenge is to reduce the number of signals required to transmit information in the network—for example , through sparse-coding schemes ( text in bold appear in the Glossary section ) , in which information is represented by the activity of a small proportion of neurons in a population [5–7] . A number of different definitions of sparsity can be found in the literature [8 , 9] , which can sometimes lead to controversy as to which codes can still be considered sparse [8] . An extreme example is the so-called local code , in which each unique event , or “context , ” is encoded by a single active neuron , or “grandmother cell” [10] ( illustrated in the left column of Fig 1A ) . Local codes not only suffer from low representational capacity , because they allow a population of N neurons to encode at most N contexts , but also require a large number of neurons to cover the space of possible contexts . On the other hand , a dense code represents each context by the combined activity of all neurons in the population ( Fig 1A , right column ) . In theory , dense codes lead to high representational capacity ( at M activity levels , allowing for MN contexts to be encoded ) , but they also suffer from neuronal cross talk because every neuron is involved in every context . Alternatively , sparse codes ( Fig 1A , center column ) can be described as a trade-off between the benefits and drawbacks of dense and local codes , in which each context is encoded by a different subset of neurons in the population . [5] . In general , sparse coding reduces the overall neural activity necessary to represent information . Another approach to address this challenge is to reduce the number of variables required to represent a particular input , stimulus , or task space , a process known as dimensionality reduction . Although responses of individual neurons are often complex and highly nonlinear , a population of neurons might share activity patterns because of individual neurons in the population not being independent of each other . Dimensionality reduction methods have proved useful in elucidating these shared activity patterns and thus effectively explaining population activity using a lower number of variables than there are neurons in the population ( for a recent review , see [14] ) . Neurons often encode several behaviorally relevant variables simultaneously [15–18] , allowing for multifaceted representations of high-dimensional stimulus spaces . For example , a population of neurons tasked with encoding human faces might opt to represent each individual face as a combination of a set of standard faces ( Fig 1B , left column ) . In such a holistic representation of faces [11] , each individual neuron would itself respond to a face as a whole ( i . e . , a face “template” ) without explicitly representing individual face components , and an arbitrary face could be represented by combining different face templates ( e . g . , by adding 10% of template 1 to 20% of template 2 and subtracting 30% of template 3 ) . On the other hand , faces can also be represented as a combination of individual face components , such as eyes , noses , and mouth , in what is known as a parts-based representation ( Fig 1B , right column ) [12 , 19] . Both approaches allow for representing arbitrary faces as a combination of neural activity but have drastically different consequences on the set of stimulus features each neuron responds to . Although visual information from the eyes , nose , and mouth would of course be included in a holistic face representation , that information would not be explicitly represented as structural units in their own right [11] . Linear combinations of holistic components often involve complex cancellations between positive and negative contributions and thus lack the intuitive meaning of adding parts to form a whole . In contrast , a parts-based representation allows for only nonsubtractive combinations of stimulus features [12] . Although the relevant stimulus dimensions are often not known a priori , several sophisticated mathematical techniques exist that allow us to discover these representations directly from experimental data [14 , 19–23] . In this article , we review evidence from experimental and theoretical studies suggesting that a number of neuronal responses can be understood as an emergent property of nonnegative sparse coding ( NSC ) , an efficient population coding scheme based on dimensionality reduction and sparsity constraints . In particular , we review evidence for NSC in sensory areas that efficiently encode external stimulus spaces , for associative areas to conjunctively represent multiple behaviorally relevant variables , and for the basal ganglia to coordinate movement .
The fundamental principle of efficient coding is that a sensory system is adjusted to the specific statistics of the natural environment from which it encodes and transmits information [24–27] . Efficiency , in this context , is an information-theoretic term that should not be confused with “minimizing energy expenditure . ” Instead , a sensory pathway is treated as a noisy communication channel , in which the goal is to maximize the rate at which information can be reliably transmitted by minimizing the redundancy between representational units . Early theories of efficient coding [24 , 25] were developed based on the visual system . Attneave [25] pointed out that there is a significant degree of redundancy in natural visual images because of correlations in both the spatial and temporal domains ( for a recent review , see [28] ) . For example , the luminance values of a pair of pixels separated by a fixed distance in a natural image are likely to be highly correlated ( Fig 2A ) . These statistical regularities constrain the images a visual system is likely to encounter to a tiny fraction of the set of all possible images . It was therefore argued that the visual system should not waste resources on processing arbitrary images but instead use statistical knowledge about its environment to represent the relevant input space as economically as possible . Extending this idea to the neural level , Barlow [24] proposed that the goal of early neurons in sensory processing is to transform raw visual inputs into an efficient representation such that as much information as possible can be extracted from them given limited neural resources . This efficient coding principle has been able to explain a wide variety of neuronal response properties in the early visual system , such as the center-surround structure of receptive fields ( RFs ) in the retina [30] , temporally decorrelated signals in the lateral geniculate nucleus ( LGN ) [31] , and the coding of natural scenes in the primary visual cortex ( V1 ) [9] . At the level of single neurons , efficient coding suggests that the information carried by a neuron's response can be maximized by using all response levels with equal frequency [29 , 32 , 33] . For example , in the case of a neuron representing a single input variable with a single output variable , information is maximized when the input–output function corresponds to the cumulative probability function for the different input levels [29] , as shown in Fig 2B . Note that this coding procedure amplifies inputs in proportion to their expected frequency of occurrence rather than reserving large portions of its dynamic range for improbable inputs [29 , 32] . On the other hand , if the input–output function sensitivity is chosen as too low , high levels of the stimulus feature will be indistinguishable as the response function saturates; if the sensitivity is set too high , low levels of the stimulus feature cannot drive responses [29] . At the level of neuronal populations , neural responses should be both decorrelated ( i . e . , independent from one another ) and sparse ( i . e . , involve only a small fraction of neurons in the population ) [27] . Taking these ideas a step further , Olshausen and Field [34] noted that natural images contain statistical dependencies beyond linear pairwise correlations among image pixels and argued that these higher-order correlations should be taken into account when developing an efficient code . Their goal was thus to find a linear coding strategy capable of reducing these higher-order forms of redundancy . Linear sparse coding is one such strategy , in which monochromatic images I ( x , y ) are described in terms of a linear superposition of a number of B basis functions , wb ( x , y ) : I ( x , y ) =∑b=1Bwb ( x , y ) hb , ( 1 ) where hb are stochastic coefficients that are different for each image [35 , 36] . Learning a sparse code for images thus involved determining the values of both wb ( x , y ) and hb for all b and ( x , y ) , given a sufficient number of observation of images , under the constraint that hb be sparse . In this context , hb was considered sparse if it took very small or very large ( absolute ) values more often than a Gaussian random variable would [36] . This sparsity constraint allowed for basis functions that were not needed to describe a given image structure to be weeded out . When Olshausen and Field applied linear sparse coding to natural images , they found that the emerging basis functions were qualitatively similar in form to RFs of simple cells in V1 [35 , 37] , thus giving empirically observed RFs an information-theoretic explanation . In this context , hb in Eq 1 corresponded to the ( signed ) activation value of a particular V1 neuron , and wb ( x , y ) were the connection weights ( or synaptic weights in an artificial neural network ) that were closely related to that neuron's RF . Sparsity , in this context , is an information-theoretic concept related to how efficiently and completely information is encoded with the basis functions described previously . Please note that this is different from empirical observations of brain areas being “sparsely” activated; that is , sparse population activity does not necessarily imply that a brain area implements a sparse-coding scheme . This confusion is fueled in part by the wide variety of definitions of sparsity used in the literature [8 , 38] . For example , even though sparse coding ( as a theoretical framework ) applied to natural images yields V1-like RFs , recent evidence suggests that neural activity in V1 might not be as sparsely activated as previously thought [39 , 40] . However , V1 still codes stimuli efficiently [40] . Olshausen and Field went on to show that the set of basis functions that best described V1 RFs was greater in number than the effective dimensionality of the input ( which they termed an overcomplete basis set ) [37] . It is worth noting that sparse coding with an overcomplete basis set is typically associated with an anatomical fan-out motif , such as expanding 1 million optic nerve fibers into more than 100 million V1 neurons or from a small number of mossy fibers to a 100-fold–larger number of granule cells in the cerebellum . However , as pointed out by Hoyer [41] , linear sparse coding falls short of providing a literal interpretation for V1 simple-cell behavior for two reasons: ( 1 ) every neuron could be either positively or negatively active , and ( 2 ) the input to the neural network was typically double-signed , whereas V1 neurons receive visual input from the LGN in the form of separated , nonnegative ON and OFF channels . In order to transform Olshausen and Field's sparse coding from a relatively abstract model of image representation into a biologically plausible model of early visual cortex processing , Hoyer [41 , 42] thus proposed to enforce both input signal and neuronal activation to be nonnegative ( though still allowing inhibitory connections ) . This seemingly simple change had remarkable consequences on the quality of the sensory representation: whereas elementary image features in the standard sparse-coding model could “cancel each other out” through subtractive interactions , enforcing nonnegativity ensured that features combined additively , much like the intuitive notion of combining parts to form a whole . The resulting parts-based representations resembled RFs in V1 much more closely than other holistic representations . These considerations led to the formulation of NSC in its current form . As a special case of linear sparse coding , NSC shares the same goal of accurately describing observed data as a superposition of a set of sparsely activated basis functions , as well as enforcing dimensionality reduction . In addition , NSC requires all basis functions and activation values ( i . e . , wb ( x , y ) and hb in Eq 1 ) to be nonnegative . However , NSC is more than just linear sparse coding with nonnegative weights . For example , whereas linear sparse coding typically uses a larger number of basis functions than there are dimensions in the input ( thus achieving dimensionality expansion ) , NSC makes use of nonnegative matrix factorization ( NMF ) to achieve dimensionality reduction . This has interesting implications for the kinds of basis functions that can be learned . Most prominently , the nonnegativity constraints used in NMF force the different basis functions to add up linearly , thus leading to the distinctive parts-based representations . Consider S observed stimuli or data samples , each composed of F observed feature values , such as a collection of S images I ( x , y ) s ( s∈[1 , … , S] ) from the previous example , each consisting of F different grayscale values . If we arrange the observed feature values of the s-th observation into a vector v→s ( i . e . , by flattening each observed image ) , and if we arrange all vectors into the columns of an F×S data matrix V , then linear decompositions describe these data as V≈WH , ( 2 ) where W is an F×B matrix that contains as its columns the B basis functions of the decomposition ( i . e . , the b-th column of W corresponding to wb ( x , y ) ∀x , y in Eq 1 ) , and H is a B×S matrix containing as its columns the activation values of each basis function for a particular input stimulus ( i . e . , the b-th column of H corresponding to hb ∀b in Eq 1 ) . The difference between V and WH is termed the reconstruction error . The goal of NSC is then to find a linear decomposition of V that minimizes the reconstruction error while guaranteeing that H is sparse . This can be achieved by minimizing the following cost function [42]: minW , H12‖V−WH‖2+λ∑ijf ( Hij ) , ( 3 ) subject to the constraints ∀ij:Wij≥0 , Hij≥0 , and ‖w→i‖=1 , where w→i denotes the i-th column of W . Here , the left-hand term describes the reconstruction error , whereas the right-hand term describes the sparsity of the decomposition . The trade-off between accurate reconstruction and sparsity is controlled by the parameter λ ( where λ≥0 ) , whereas the form of f defines how sparsity is measured ( a typical choice is the L1 norm on H ) . Analogous to efficient coding , Eq 3 forces prediction errors to be amplified in proportion to their expected frequency of occurrence because a more frequent event would show up more frequently in V . Hence , accounting for a rare observation at the expense of ignoring a more common one would result in an increased reconstruction error . In the case of λ = 0 , Eq 3 reduces to the squared-error version of NMF . Although NMF enforces all elements of W and H to be nonnegative , the resulting decomposition might not be sparse , depending on the number of basis functions B . In order to emphasize decompositions in which H is sparse , Eq 3 should be minimized with λ>0 [42] . Another open parameter is the number of basis functions , B , which controls the predictive power of the model and must be determined empirically . With a small number of basis functions , NSC is unlikely to achieve a low reconstruction error , be it in familiar contexts ( training data ) or in novel contexts ( held-out test data ) . In this case , the error depends on the systematic bias of the model , and the model is said to underfit the data ( left-hand side of Fig 3 ) . With increased model complexity , the model can learn subtle differences between different contexts with high accuracy , leading to a reduced bias ( training ) error . However , with increased complexity , the model is more likely to learn patterns between training contexts that arise either from underlying noise or from spurious correlations . As a result , the model will respond according to these learned patterns when a novel context is presented ( rather than according to the underlying actual relationships ) , in which case the model is said to overfit the data ( right-hand side of Fig 3 ) . Hence , the goal of a successful model is to find the ideal compromise in the bias–variance error trade-off [43] ( labeled “best model” in Fig 3 ) . Analogously to [35 , 37] , the basis functions obtained in NSC can be interpreted as the connection weights of a population of simulated neurons in an artificial neural network . In other words , under NSC , the number of basis functions B corresponds to the number of output neurons , and the response of the b-th model output neuron ( b∈[1 , … , B] ) to a particular input stimulus s , termed rbs , can be computed by feeding the dot product of that neuron's connection weights ( i . e . , the b-th column in W , w→b ) and a data vector ( i . e . , the s-th column in V , v→s ) to an activation function Θ: rbs=Θ ( w→b⋅v→s ) , ( 4 ) where “⋅” denotes the dot product . For example , the linear response of a model neuron can be calculated by setting Θ to the identity function Θ ( x ) = x . Note that the response of the model neuron to different stimuli s∈[1 , … , S] involves different columns of V but always relies on w→b . Thus , we can utilize W ( which must remain fixed once learned ) and Eq 4 to simulate a model neuron's response to arbitrary input stimuli by replacing the column in V with new input . This allows us to investigate the response properties of individual model neurons much in the same way that experimental neuroscientists study biological neurons . This is important because it means that NSC can be used to model neural activity in the brain , and the resulting activity patterns generated by NSC can be compared to and evaluated against experimental findings . It is important to note that the absence of negative weights in Eqs 2–4 does not preclude the modeling of inhibitory connections or even posit that inhibitory connections cannot participate in NSC . Rather , one important aspect of NSC is the parts-based , NMF-like decomposition of V; one way to achieve this is by enforcing nonnegativity constraints on W and H . Several studies have successfully incorporated inhibitory connections into their NSC-based models . One approach is to model them as nonnegative synaptic conductances . For example , Hoyer [41] used NSC to model V1 neurons as receiving input from both excitatory ON and inhibitory OFF cells in the LGN . Using prewhitened natural images , Hoyer sampled 12×12 pixel patches from the images and then separated positive and negative values into separate channels . Each image patch was thus represented by a 2×12×12 = 288 dimensional vector , each element of which mimicked the activity of an ON or OFF cell in response to the image patch . These vectors were then arranged into the columns of V . This procedure not only preserved the parts-based quality of the encoding but also allowed the modeling of the convergence of ON and OFF pathways . Another approach is to drop the nonnegativity constraint on W and thus effectively operate with both positive and negative synaptic weights . Only recently did it become clear that this approach was able to preserve the parts-based quality of the encoding ( as long as nonnegativity of H was enforced ) [44] , thus simplifying the construction of more complex network topologies .
The notion of parts-based object recognition is compatible with hierarchical models of vision , in which activation of simple features feeds into the activation of complex features [51] . There is a long history of debate as to whether humans detect faces based on their individual parts or as correctly arranged wholes ( for reviews , see [11 , 52 , 53] ) . The working hypothesis is that the brain might use holistic face information as an early gating mechanism to allow visual stimuli access to the face processing module but that most cortical circuitry relies on parts-based information [53] . Converging evidence from human imaging studies and primate physiology suggests that faces are processed in localized “patches” within IT [54] , where cells detect distinct constellations of face parts [55 , 56] , such as eyes [57] , and that whole faces can be recognized by taking linear combinations of neuronal activity across IT [19 , 58] . An influential paper by Lee and Seung [13] found that applying NMF to a database of face images yielded sparse , localized features that resembled parts of a face ( Fig 4A ) in a similar fashion to responses in area IT . In their case , NMF acted on an F×S data matrix V , whose rows corresponded to distinct features of the input ( e . g . , F different pixels of an image ) and whose columns corresponded to different stimuli or observations of those features ( e . g . , S different images ) . NMF was used to decompose the matrix into two reduced-rank matrices ( Fig 4 , inset ) whose linear combination could be weighted such that the product of W and H provided an accurate reconstruction of V ( see Eq 2 ) . A particular image , in this case encoded by F = 19×19 = 361 pixels could be accurately represented by a linear combination of a small number ( B = 49 ) of encoding variables or “basis images” ( Fig 4A ) . Such a representation is reminiscent to neural processing in IT , an area in the ventral visual “what” stream involved in encoding high-level object identity [58 , 59] , in which images of whole faces can be linearly reconstructed using responses of approximately 200 neurons that each respond to a certain set of physical facial features [19] . Interestingly , such a parts-based representation is not specific to face processing in IT; the same principle can be extended to body-selective regions in IT [60 , 61] . Although there seems to be a consensus that information-theoretic explanations are relevant when investigating early sensory areas , higher-order brain areas are often considered to be specialized for performing tasks ( e . g . , recognizing objects , making decisions , navigating an environment ) rather than the efficient encoding of information . It is therefore possible that the essential components of NSC might well be present in higher-order areas but , to date , have gone unnoticed . Because of its roots in efficient coding theories of natural image processing , NSC figures prominently in the vision neuroscience literature . For example , NMF-based models were able to reconstruct in vitro neuronal spike trains from the salamander retina [44 , 62] . By combining spike-triggered average with NMF , Liu and colleagues [44] were able to identify the subunit layout of retinal ganglion cells ( Fig 5 ) . This technique , termed spike-triggered NMF ( STNMF ) , involved applying NMF to the collection of those stimulus patterns contained in a spatiotemporal white-noise sequence that caused a given neuron to spike . Akin to common reverse-correlation analysis , the researchers averaged the collection of spike-eliciting stimulus segments to form the spike-triggered stimulus ensemble ( Fig 5A ) . STNMF then decomposed the ensemble of effective spike-triggered stimuli into a matrix W containing a set of modules ( or basis functions ) and a matrix H containing a set of hidden coefficients . Intuitively , the modules derived by STNMF should capture the subunit decomposition of the cell's RF because the spike-eliciting stimuli should have essential statistical structure imprinted on them by the subunits , such as correlations between pixel values [44] . And indeed , the identified modules corresponded to individual presynaptic bipolar cells , as verified by multielectrode array recordings with simultaneous recordings from individual bipolar cells through sharp microelectrodes [44] . This allowed the researchers to improve predictions about how ganglion cells respond to natural stimuli without the need to guess a specific model structure that may be constrained in terms of the size , shape , number , or nonlinearity of ganglion cell subunits . NSC has been extensively applied to early visual cortex , where it has successfully explained orientation and frequency tuning of simple and complex cells in V1 [41] as well as edge-like pooling of spatial frequency channels in V2 [63] , including RF properties such as end-stopping and contour integration [64] . These theoretical findings are in good agreement with a large body of research documenting the sensory response of V1 across animal models ( e . g . , [65–68] ) , although they are not without controversy . For example , one study [67] criticized that some of the early sparse-coding models generated RFs that looked like stereotyped edge detectors and did not capture the diversity of RF structure observed in cat and monkey V1 . However , by adjusting these models to limit the number of active neurons ( “hard” sparsity ) instead of limiting mean neuronal activity ( “soft” sparsity ) , Rehn and Sommer [69] were able to account for the diversity of shapes in biological RFs . Other researchers were concerned that the apparent sparse activation of V1 was an artifact of using simple artificial stimuli such as sinusoidal gratings and drifting bars , but Vinje and Gallant [9] were able to show that natural viewing conditions actually increased the sparsity of V1 activation . However , a number of recent studies suggest that responses are neither sparse nor low dimensional in V1 of the mouse [39 , 40] and monkey [70] . Using high-density electrophysiology , Stringer and colleagues [40] found that the response of more than 10 , 000 visual cortical neurons to 2 , 000 image stimuli is high dimensional . In monkey V1 , one needs to look at many principal components to decode natural images , and these principal components reflect contributions from most of the recorded neurons [70] . In addition , V1 neurons in the mouse might encode both visual stimuli and behavior in a mixed representation: a recent study found no separate sets of neurons encoding stimuli and behavioral variables , but each neuron multiplexed a unique combination of sensory and behavioral information [39] . These findings suggest that efficient coding might render an incomplete picture of sensory processing in V1 and that more studies are needed to reevaluate past findings . To this end , Stringer and colleagues [40] suggested that the population code of visual cortex might be determined by two constraints: efficiency , to make best use of the limited number of neurons , and smoothness , which allows similar stimuli to evoke similar responses . In summary , there is a large body of research showing that computational models based on efficient coding , such as NSC , can account for a variety of response properties in early visual cortex . Although methods like spike-triggered average [71] and dimensionality reduction [72] give us confidence that we have a good understanding of the sensory response in V1 , this understanding remains far from complete [73 , 74] and in fact might be missing a number of dimensions related to task , state , or behavior [39 , 40] . With the exception of face processing in IT [13 , 19] , NSC has yet to be applied to higher-order areas in the ventral visual pathway . The success of NSC in explaining V1 and V2 response properties suggests that it might be possible to extend the model to texture integration in V4 . Our group found evidence for NSC in the dorsal subregion of the medial superior temporal ( MSTd ) area [46] , which is part of the visual motion pathway in the dorsal visual stream . Neurons in MSTd respond to relatively large and complex patterns of retinal motion ( “optic flow” ) , owing to input from direction- and speed-selective neurons in the middle temporal ( MT ) area ( for a recent review , see [75] ) . Although MSTd had long been suspected to be involved in the analysis of self-motion , the complexity of neuronal response properties has made it difficult to experimentally investigate how neurons in MSTd might perform this function . When our group applied NMF to simulated neural activity patterns whose statistical properties resembled that of experimentally recorded MT neurons [46] , we found a sparse , parts-based representation of retinal flow ( Fig 4B ) similar to the parts-based representation of faces encountered by Lee and Seung [13] . The resulting “basis flow fields” showed a remarkable resemblance to RFs of MSTd neurons , as they responded to an intricate mixture of 3D translational and rotational flow components in a subset of the visual field . As a result , any flow field possibly to be encountered during self-movement through a 3D environment could be represented by only B = 64 simulated MSTd neurons , as compared with F = 9 , 000 simulated MT input neurons . This led to a sparse and parts-based population code in which any given stimulus could be represented by only a small number of simulated MSTd neurons [46] . Fig 6 shows the distribution of direction preferences of MSTd-like model units ( Fig 6A and 6B; [46] ) for rotation and translation , respectively . Each data point in the scatter plots specifies the preferred 3D direction of a model unit . Histograms along the boundaries show the marginal distributions of azimuth and elevation preferences . Not only did individual units match response properties of individual neurons in macaque MSTd [76] ) , but the model was able to recover statistical properties of the MSTd population as a whole , such as a relative overrepresentation of lateral headings . MSTd is known to encode a number of perceptual variables , such as the direction of travel ( heading ) and eye rotation velocity . During forward movement , retinal flow radiates out symmetrically from a single point , the focus of expansion ( FOE ) , from which heading can be inferred . However , instead of consisting of a set of distinct subpopulations , each specialized to encode a particular perceptual variable , MSTd has been found to consist of neurons that act more like basis functions , in which a majority of cells were involved in the simultaneous encoding of multiple perceptual variables ( Fig 6C ) . A similar picture emerged when we investigated the involvement of MSTd-like model units in the encoding of both heading and eye rotation velocity ( Fig 6C ) . Interestingly , the sparsity regime in which model MSTd achieved the lowest heading prediction error ( Fig 6D ) was also the regime in which MSTd-like model units reproduced a variety of known MSTd visual response properties ( for experimental details , refer to [46] ) . In contrast to findings about early visual cortex , this regime does not use an overcomplete basis set [35] , yet it can still be considered a sparse coding regime [8] because only a few MSTd-like model units were needed to recover the stimulus , and each model unit responded to a subset of stimuli ( see Fig 8C in [46] ) . Such an intermediary sparse code might be better suited ( as opposed to an overcomplete basis set ) for areas such as MSTd because the increased memory capacity of such a code might lead to compact and multifaceted encodings of various perceptual variables . Taken together , the computational modeling work on MSTd described previously suggests that NSC is not specific to primary sensory areas and may be observed in other downstream sensory regions . Analogous to early visual cortex , the auditory system is believed to decompose auditory signals into a set of elementary acoustic features [77] such that the complete acoustic waveform can be described by a sparse population code that operates near an information-theoretic optimum [77–79] . It is therefore not surprising that computational models based on NSC have been very successful at describing the spectro-temporal RF of neurons in the primary auditory cortex ( A1 ) [80 , 81] . Response properties of A1 neurons are well described by a spectrogram; they are often tuned to stimulus frequency but are rarely phase locked to oscillations of the sound waveform [82] . The cortical representation of auditory signals seems to not only be sparse but also rely on statistically independent acoustic features [83] . Similar to visual cortex , auditory cortex is hierarchically organized , with neurons in A1 responding to simple acoustic features of natural sounds and higher-order areas responding to more behaviorally relevant stimuli . The anterior superior temporal region of auditory cortex , for example , responds to categories of acoustic objects , such as sounds produced by voices and musical instruments [82] . An intriguing question for future modeling studies is therefore whether NSC can be extended to the next level of the auditory hierarchy: Would it be possible to construct more complex acoustic objects from a sparse , parts-based set of elementary , A1-like acoustic features ? And would the representation of such acoustic objects resemble neuronal responses in the anterior superior temporal region of auditory cortex ? Taken together , we suggest that auditory cortex is a good example for efficient and NSC-based coding in a sensory system other than the visual cortex , in which further study is warranted . The olfactory cortex is another nonvisual cortical area worth investigating for NSC-like responses . In contrast to most other sensory modalities , the basic perceptual dimensions of olfaction remain unclear . In particular , the olfactory modality is intrinsically high dimensional and lacks a simple , externally defined basis analogous to wavelength or pitch on which elemental odor stimuli can be quantitatively compared ( for a recent review , see [84] ) . Odors evoke complex responses in granule cells ( located in the olfactory bulb ) that evolve over hundreds of milliseconds [85] . Granule cells use a sparse combinatorial code to convey information about odor identity and concentration [86 , 87] . Downstream from the olfactory bulb , odors tend to activate a small but consistent proportion ( approximately 10% ) of cortical neurons in the piriform cortex [88] , which is thought to form odor object percepts [89 , 90] . Although piriform cortex is not topographically organized , a spatial structure can be discerned when examining the projections of output neurons , which are highly segregated and functionally specific . Whereas the anterior piriform cortex is associated with the encoding of odor identity and odor structure , the posterior piriform cortex is involved in associational aspects of odors , such as valence and similarity [89 , 91] . A compelling piece of evidence for NSC in the olfactory system was recently provided by Castro and colleagues [48]: In an effort to elucidate the dimensions along which perceptual space might be organized in the olfactory system , they applied NMF to a perceptual dataset built from 144 monomolecular odors , each represented by a 146-dimensional vector ( an “odor profile” ) . Each dimension in the odor profile corresponded to the rated applicability of a number of semantic labels , such as “sweet , ” “floral , ” and “heavy . ” By applying NMF to the odor profile , they showed that a set of 10 sparsely activated basis functions could accurately describe any odor in the dataset ( Fig 7A ) . Interestingly , NMF revealed a prominent block diagonal structure to the full matrix H ( Fig 7B ) , indicating that ( 1 ) a given odor tended to be characterized by a single prominent basis function , implying that the basis functions recovered by NMF were perceptually meaningful , and ( 2 ) all ten basis functions were being used approximately with equal frequency , implying that the basis functions recovered by NMF could span the space of behaviorally relevant odors . This suggests that a given odor percept may be considered an instance of one of several fundamental qualities . Furthermore , NMF recovered basis functions whose descriptors aligned with perceptual dimensions highlighted in several previous analyses of odor space , including but not limited to relative pleasantness ( e . g . , “fragrant , ” “sickening” ) and potential palatability ( “woody , resinous , ” “chemical , ” “sweet , ” and “lemon” ) . Odors clustered predominantly along these axes ( as illustrated in Fig 7C ) for three specific basis functions [48] . In summary , although sensory processing in the olfactory system remains an area of active research , there is evidence consistent with a sparse and parts-based encoding of odor identity and concentration . Only recently have NSC-based methods been employed to elucidate the neural code for olfaction . Future studies may provide additional supporting evidence . In early areas of primary somatosensory cortex ( S1 ) , a number of parallels can be drawn to sparse , reduced information processing observed in other primary sensory cortices . First , activity in rodent barrel cortex , a region of S1 that is a major target for somatosensory inputs from the whiskers via the thalamus , can be extremely sparse [92–94] , similar to activity in A1 . Consequently , sparse-coding models have successfully explained the response properties of individual neurons in rat barrel cortex ( e . g . , Hafner and colleagues [95] ) . Second , similar to V1 , neurons in primate areas 3b and 1 of S1 act like Gabor filters for tactile orientation [96 , 97] . The same is true for rat barrel cortex [98] . Third , similar to visual area MT , primate S1 contains a subpopulation of neurons that can infer the direction of tactile motion from a spatiotemporal pattern of activation across a 2D sensory sheet ( i . e . , the skin ) [99] . Specifically , neurons in area 1 of S1 tend to respond to plaid textures in the same fashion that MT neurons respond to visual plaids [99] . These findings suggest that much of what can be said about sparse and parts-based information processing in visual cortex also applies to S1 . One NSC-like model that has enjoyed success in explaining complex S1 rodent response properties is the rectified latent variable model ( RLVM ) , a combination of nonlinear dimensionality reduction with nonnegativity constraints . In an effort to elucidate the stimulus dimensions that individual S1 neurons respond to , Whiteway and Butts [100] applied RLVM to a two-photon imaging dataset of hundreds of simultaneously recorded neurons in mouse barrel cortex while the animal was performing a tactile discrimination task . Interestingly , they found basis functions that properly identified individual neurons . Similar to the recorded neuronal responses , these basis functions were closely related to both the tactile stimulation as well as nonstimulus aspects of the behavioral task . Furthermore , RLVM achieved a lower reconstruction error than other linear dimensionality reduction techniques such as principal component analysis ( PCA ) , thus highlighting the benefit of using NMF-based decompositions over PCA to explain neural data . However , NSC has not been observed in nonhuman primate somatosensory cortex . Tactile information from various submodalities converges at later stages of monkey S1 [101 , 102] and is multiplexed across different time scales using both rate and spike timing codes [103] . These regions might represent different stages in the processing pipeline leading to form and texture perception [104] . Primate area 2 of S1 is known to integrate both tactile and proprioceptive stimuli; for example , some neurons respond only to active reaching movements , some respond only to passive movements ( e . g . , unexpected perturbations to the hand that generate passive limb displacements ) , and others respond to both [105] . These complex response properties may argue against a sparse and parts-based code in area 2 . Taken together , neurons in early somatosensory cortex respond to a small number of stimulus dimensions , not unlike to sensory neurons in early visual and auditory cortex . However , current evidence argues against NSC in higher areas of somatosensory cortex . The parallels to the visual system are striking though: area 1 , which resembles visual area MT by showing Gabor-like responses to tactile motion , feeds into area 2 , which resembles visual area MSTd by showing intermingling of responses to tactile and proprioceptive stimuli ( analogous to intermingling of visual and vestibular stimuli in MSTd ) . It is therefore not unthinkable that an NSC-like model that operates on neuronal inputs to area 2—constructed analogous to [46]—could reproduce some of these response properties . However , until the neuronal mechanisms underlying these complex response properties are better understood , one would have to conclude that NSC might not apply to later stages of somatosensory cortex . In our own work , we found evidence that NSC can explain response properties in RSC , an area important for navigation and spatial memory [106–108] . Neurons in the RSC conjunctively encode multiple variables related to the environment and one's position and movement within it ( e . g . , position , head direction , linear velocity , and angular velocity ) , allowing the representation of spatial features of the environment with respect to multiple reference frames [109] . Using a similar methodology to [46] , we applied NMF , with a sparsity constraint , to parameterized behavioral variables extracted from electrophyisiological recordings of RSC neurons in the rat [109] while the animal ran back and forth on a W-shaped track ( for experimental details , see Supporting information ) . We found a sparse and parts-based representation for behaviorally relevant variables such as the animal's position , head direction , and movement direction ( Fig 4C ) . Interestingly , model RSC neurons encoded these variables with respect to multiple frames of reference ( e . g . , head direction: allocentric reference frame , linear velocity: route-based reference frame ) . The dimensionality of the stimulus space was drastically reduced from F = 417 input neurons to a set of B = 30 model RSC neurons . The basis functions recovered by NMF were then used to generate simulated responses of model RSC neurons according to Eq 4 , and the simulated responses were compared with neuronal responses from the electrophysiological recordings . Interestingly , simulated neuronal activity could be classified into three broad categories , with remarkably similar population statistics to rat RSC: ( 1 ) responding to left and right turns on a specific position along the route , ( 2 ) responding to left and right turns regardless of the position along the route , and ( 3 ) exhibiting complex and robust firing patterns without turn sensitivity ( see Fig 8A and 8B as well as Supporting information ) . Taken together , this study suggests that neuronal population activity in RSC is consistent with NSC . This is an example that NSC can apply outside sensory cortex , even where responses have not traditionally been considered sparse or parts based . There is computational evidence for a reward-driven variant of NSC in the basal ganglia , a cluster of deep forebrain nuclei that are involved in the processing of motor , associative , and limbic information ( for recent reviews , see [3 , 110] ) . The basal ganglia network may be viewed as multiple parallel loops where cortical and subcortical projections interact with internal reentral loops , forming a complex network ideally designed for selecting and inhibiting simultaneously occurring events and signals ( for a recent review , see [111] ) . To achieve this function , the basal ganglia connect most cortical areas to the frontal cortex through a series of convergent and sparsely connected pathways [112] , in which signals from tens of millions of cortical neurons are projected onto a 10–10 , 000-fold smaller population of neurons in different subnuclei of the basal ganglia [3] . Similar to the convergence of 100 million photoreceptors onto 1 million optic nerve fibers in the retina , these highly convergent pathways from cortex to the basal ganglia suggest a potential role for dimensionality reduction . One possible model , termed the reinforcement-driven dimensionality reduction ( RDDR ) model , suggests that dimensionality reduction in the cortico-basal ganglia pathway is achieved via a combination of Hebbian and anti-Hebbian learning rules that are implemented by feedforward excitatory and lateral inhibitory connections [3 , 49] . These learning rules control the strength of synaptic weights in the network by altering the weight of a given synapse in proportion to the correlation between the firing rates of its presynaptic and postsynaptic neurons . In Hebbian learning , synaptic weights are strengthened given a positive correlation ( leading to a phenomenon referred to as long-term potentiation [LTP] ) , whereas synaptic weights are depressed if the firing rate correlation is negative ( leading to long-term depression [LTD] ) . On the other hand , in anti-Hebbian learning , which is typically applied to inhibitory connections , correlated activities are subjected to LTD , and uncorrelated activities are subjected to LTP . In order to implement dopamine-modulated Hebbian learning in this model , a reinforcement signal was used to dictate the level of dopamine in the circuit ( 1 for reward-related events , 0 for the absence of reward-related events , and negative values to simulate dopamine depletion ) [49] . The value of the reinforcement signal then determined the sign and magnitude of each synaptic weight change . In the RDDR model , a reinforcement signal corresponding to dopamine modulates the Hebbian learning rule of the feedforward projections , allowing the network to learn to extract input dimensions that are associated with reward activity while suppressing behaviorally irrelevant input dimensions . Whereas the original RDDR model was a neural network–based model for performing PCA [49] , later iterations incorporated nonnegativity constraints on the connection weights that effectively transformed the model into an NMF variant [3] . The model predicted that these lateral connections facilitated learning by shaping correlations between neurons in the corticostriatal projections using dopamine-modulated LTP and LTD , which has yet to be experimentally validated . In addition to suggesting a role for lateral connectivity in the basal ganglia , the RDDR model also advanced understanding of basal ganglia dysfunction in movement-related disorders such as Parkinson’s and Huntington’s disease . Previous studies had indicated that lesions to functionally healthy basal ganglia had minimal impact on behavior . Bar-Gad and colleagues [49] then suggested that this was an expected finding because of the network's ability to reorganize connections , whereas abnormal dopamine levels should significantly alter the reinforcement signal that controls the model's ability to discriminate behaviorally relevant input signals ( as in Parkinson disease ) . Accordingly , restoration of background dopamine levels via dopamine replacement therapy alleviates the symptoms , consistent with results of dopamine depletion and restoration in the model . In summary , NSC is a prime candidate to allow the basal ganglia to compress information in the cortico-basal ganglia pathway and extract input dimensions that are associated with reward activity . However , the complexity of the basal ganglia network has so far prohibited a deep scientific understanding of the multifaceted neural computations it performs .
We reviewed compelling evidence that a wide range of neuronal responses can be understood as an emergent property of efficient coding due to dimensionality reduction and sparsity constraints . In particular , NSC might be employed by sensory areas to efficiently encode external stimulus spaces , by some associative areas to conjunctively represent multiple behaviorally relevant variables , and possibly by the basal ganglia to coordinate movement . NSC is tightly connected to a number of unsupervised learning techniques , such as NMF ( a popular tool for high-dimensional data analysis [113] ) , k-means clustering ( an algorithm used to partition n observations into k clusters [114] ) , and independent component analysis ( ICA ) ( a computational method for separating a multivariate signal into additive , statistically independent subcomponents ) . Both NMF and ICA are capable of decomposing high-dimensional data into parts-based representations—in contrast to PCA , which usually results in holistic representations [13] . As originally noted by Hoyer [42] , if the fixed-norm constraint is placed on the rows of H instead of the columns of W , Eq 3 can be directly interpreted as the joint log-posterior of the basis functions and hidden components in the noisy ICA model [64] . Similarly , NSC is closely related to compressed sensing ( for a recent review , see [115] ) , and a recent study has even suggested to combine the two [116] . Compressed sensing posits that neurons might implement dimensionality reduction by randomly projecting patterns of activity into a lower-dimensional space—namely , by synaptically mapping N upstream neurons to a downstream region containing M<N neurons . Analogously , compressed sensing supports dimensionality expansion by projecting into a larger downstream area [115] . The theory of compressed sensing then provides the mathematical tools to reconstruct the original space from the random projections . NSC , ICA , and compressed sensing often make similar predictions that only slightly differ in the nature of the basis function representation necessary to achieve optimal reconstruction ( for details , please refer to the Discussion of [115] ) . For example , whereas ICA emphasizes the statistical independence of unmixed sources , and compressed sensing requires basis function to be “maximally incoherent”[115] , NSC does not make any such assumptions as long as the basis functions are nonnegative . There are several nonsensory areas that may demonstrate NSC . In this section we point to evidence that suggests this is the case but also discuss how sparse activity in these regions differs from NSC in sensory systems . We suggest that further studies should be carried out to assess the potential for NSC in these regions . In addition to the areas highlighted previously , the essential components of NSC might be present in other brain regions not traditionally associated with the efficient encoding of information . We offer three testable predictions of this theory: First , we suggest that a variety of neuronal response properties can be understood as an emergent property of efficient population coding based on dimensionality reduction . Depending on input stimulus and task complexity , we expect the dimensionality of the population code to be adjusted according to the bias–variance dilemma ( Fig 3 ) . This point of operation might differ across brain areas—for example , favoring neurons that respond to a small number of stimulus dimensions in V1 [35] but giving rise to mixed selectivity in higher-order brain areas such as MSTd [46] and RSC [50 , 164] . Second , we predict that parts-based representations can explain RFs of neurons in a variety of sensory and associative cortices , including but not limited to those brain areas discussed here . In agreement with the literature on basis function representations [18 , 159 , 160] , we expect parts-based representations to be prevalent in regions where neurons exhibit a range of tuning behaviors [46] , display mixed selectivity [165 , 166] , or encode information in multiple reference frames [50 , 109 , 164] . Third , where such representations occur , we expect the resulting neuronal population activity to be relatively sparse in order to encode information both accurately and efficiently . As noted previously , sparse codes offer a trade-off between dense codes ( in which every neuron is involved in every context , leading to great memory capacity but suffering from cross talk among neurons ) and local codes ( in which there is no interference but also no capacity for generalization ) .
In conclusion , there is increasing evidence that NSC can account for neuronal response properties in a number of sensory and associative cortices , as well as subcortical areas such as the basal ganglia . Although NSC might not apply to all brain areas—for example , motor or executive function areas—the success of NSC-based models , especially in sensory areas , warrants further investigation for neural correlates in other regions .
The software used to generate some of the data presented in Figs 1B , 2 and 6A is archived on Zenodo ( 10 . 5281/zenodo . 2641351 ) . The latest version is available on GitHub: https://github . com/mbeyeler/2019-nonnegative-sparse-coding . | Brains face the fundamental challenge of extracting relevant information from high-dimensional external stimuli in order to form the neural basis that can guide an organism's behavior and its interaction with the world . One potential approach to addressing this challenge is to reduce the number of variables required to represent a particular input space ( i . e . , dimensionality reduction ) . We review compelling evidence that a range of neuronal responses can be understood as an emergent property of nonnegative sparse coding ( NSC ) —a form of efficient population coding due to dimensionality reduction and sparsity constraints . | [
"Abstract",
"Introduction",
"Nonnegative",
"sparse",
"coding",
"as",
"a",
"modern",
"variant",
"of",
"the",
"efficient",
"coding",
"hypothesis",
"Empirical",
"evidence",
"for",
"NSC",
"in",
"the",
"brain",
"Discussion",
"Conclusion",
"Data",
"availability"
] | [
"medicine",
"and",
"health",
"sciences",
"neurochemistry",
"face",
"chemical",
"compounds",
"brain",
"social",
"sciences",
"neuroscience",
"learning",
"and",
"memory",
"organic",
"compounds",
"hormones",
"face",
"recognition",
"perception",
"review",
"cognitive",
"psychology",
"cognition",
"computational",
"neuroscience",
"memory",
"amines",
"vision",
"neurotransmitters",
"catecholamines",
"coding",
"mechanisms",
"dopamine",
"animal",
"cells",
"basal",
"ganglia",
"chemistry",
"visual",
"cortex",
"head",
"biochemistry",
"cellular",
"neuroscience",
"psychology",
"cell",
"biology",
"anatomy",
"organic",
"chemistry",
"neurons",
"biogenic",
"amines",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"sensory",
"perception",
"computational",
"biology",
"physical",
"sciences",
"cognitive",
"science"
] | 2019 | Neural correlates of sparse coding and dimensionality reduction |
Chronic viral infections lead to persistent CD8 T cell activation and functional exhaustion . Expression of programmed cell death-1 ( PD-1 ) has been associated to CD8 T cell dysfunction in HIV infection . Herein we report that another negative regulator of T cell activation , CD160 , was also upregulated on HIV-specific CD8 T lymphocytes mostly during the chronic phase of infection . CD8 T cells that expressed CD160 or PD-1 were still functional whereas co-expression of CD160 and PD-1 on CD8 T cells defined a novel subset with all the characteristics of functionally exhausted T cells . Blocking the interaction of CD160 with HVEM , its natural ligand , increased HIV-specific CD8 T cell proliferation and cytokine production . Transcriptional profiling showed that CD160−PD-1+CD8 T cells encompassed a subset of CD8+ T cells with activated transcriptional programs , while CD160+PD-1+ T cells encompassed primarily CD8+ T cells with an exhausted phenotype . The transcriptional profile of CD160+PD-1+ T cells showed the downregulation of the NFκB transcriptional node and the upregulation of several inhibitors of T cell survival and function . Overall , we show that CD160 and PD-1 expressing subsets allow differentiating between activated and exhausted CD8 T cells further reinforcing the notion that restoration of function will require multipronged approaches that target several negative regulators .
Mounting evidence supports the notion that CD8 T cells contribute to the control of HIV viral replication [1] . The emergence in early infection of viral variants bearing escape mutations within sequences targeted by HIV-specific T lymphocytes is consistent with CD8 T cells exerting selective antiviral pressure [2]–[3] . However , HIV viral replication outpaces the adaptive immune response leading to the establishment of chronic infection partly because CD8 T cells are progressively deleted [4] and/or become dysfunctional [5]–[6] . Functionally exhausted T cells were originally described in a murine model of acute and chronic lymphocytic choriomeningitis virus ( LCMV ) infection whereby virus-specific CD8 T cells persist but lack effector function [7] . Exhausted CD4 and CD8 T cells have since been described in cancer [8] and chronic viral infections such as SIV [9] , HIV [10]–[12] , Hepatitis C ( HCV ) [13]–[14] and Hepatitis B ( HBV ) [15] . Functional impairment of antigen-specific responses has been shown to occur in a stepwise and hierarchical manner with proliferation , IL-2 and TNFαproduction being the first functions lost followed by IFNγ [16]–[18]; eventually cells die by apoptosis [19] . Of note , in chronic LCMV infection inhibitory receptors such as PD-1 , LAG-3 , CD160 , CTLA-4 , 2B4/CD244 , GP49 and PirB are all upregulated on exhausted LCMV-specific CD8 T cells compared to functional effector or memory cells [20]–[21] . Blocking the interaction of PD-1 and LAG-3 with their natural ligands restored virus specific T cell proliferation and effector cytokine production ( IFNγ , TNFα and CD107a ) in this mouse model . However , functional restoration was only partial suggesting the involvement and cooperation of several negative regulatory pathways [20] in the programming of T cell exhaustion . Several mechanisms that lead to antigen-specific CD8 T cell dysfunction in HIV infection have also been described; they include the lack of CD4 help [22] as well as the upregulation on HIV-specific and total T cells of several negative regulators of T cell activation including PD-1 , CTLA-4 , CD160 , 2B4 and Tim-3 . PD-1 [23]–[25] , CTLA-4 [26] and Tim-3 [27]–[28] have also been associated to HCV and HIV-specific CD4 and CD8 T cell dysfunction as the expression levels of these molecules correlated positively with plasma viral load and negatively with absolute CD4 T cell counts while there expression declined in subjects treated with highly active antiretroviral therapy ( HAART ) [23] , [26]–[27] . More recently , studies have shown in cancer [29] and HIV [30]–[31] that the co-expression of several immune inhibitory molecules on antigen-specific CD8 T cells leads to a more severe dysfunction . Of note , expression of these molecules is a by-product of T cell activation as they play an important role in T cell homeostasis . Therefore , the mechanisms that determine the functions of these molecules in T cell activation and in functional T cell exhaustion have been difficult to decipher , as most of these molecules are also upregulated upon T cell activation . We therefore analyzed the expression and function of CD160 and PD-1 on CMV and HIV-specific CD8 T cells during different stages of infection and identified 4 functionally distinct subsets of CD8 T cells ( CD160−PD-1− , CD160−PD-1+ , CD160+PD-1− , CD160+PD-1+ ) . Our data identified a unique CD160+PD-1+ subset at an advanced stage of exhaustion mostly found during chronic HIV infection ( CHI ) .
Microarray analysis of sorted CMV and HIV-specific CD8 T cells in 27 HIV-infected subjects , showed a significant increase ( 1 . 7 fold and p<0 . 05 ) in levels of expression of CD160 mRNA in HIV-specific CD8 T cells when compared to CMV-specific CD8 T cells ( Figure S1 ) . To confirm the results generated by microarray , we measured the levels of CD160 expression on CMV and HIV-specific CD8 T cells as well as total CD8 T cells isolated from 7 HIV-uninfected individuals and 38 HIV-infected subjects divided into 4 groups: acute HIV infection ( AHI; n = 7 ) , chronic/progressing infection ( CHI; n = 9 ) , successfully treated and aviremic individuals ( ST; n = 12 ) and Elite controllers ( ECs; n = 10 ) ( Table 1 , 2 ) . HIV and CMV-specific MHC class I tetramers were used to identify these cells . We also measured the expression of PD-1 on cells from individuals in these different groups as this molecule has also been shown to be upregulated on HIV-specific T cells as well as total CD8 T cells in HIV infection . Results illustrated in Figure S2A and S2B showed that significantly higher levels of CD160 , as monitored by Mean Fluorescence Intensity ( MFI ) , were found on HIV-specific CD8 T cells when compared to CMV-specific CD8 T cells in CHI subjects ( CMV: MFI = 1817; HIV: MFI = 4636; P = 0 . 0001 ) and ECs ( CMV: MFI = 2428; HIV: MFI = 3610; P = 0 . 009 ) ( Figure S2B; left panel ) . As well , the frequency of HIV-specific CD160+ T cells was significantly higher than that of CD160+ CMV-specific T cells in CHI ( CMV = 42 . 5%; HIV = 70 . 7%; P = 0 . 0001 ) and ECs ( CMV = 56 . 1%; HIV = 74 . 1%; P = 0 . 009 ) ( Figure S2B; right panel ) . The increased expression of CD160 on HIV-specific CD8 T cells was observed only during the chronic phases of HIV infection . Importantly , HIV and CMV-specific CD8 T cells expressed similar levels of CD160 in the acute phase of infection . The MFI and frequencies of CD160 on antigen-specific CD8 T cells did not differ between ST and ECs . As shown previously [23]–[24] , the frequency of PD-1+ HIV-specific T cells as well as the levels of expression of this molecule were greater on HIV than CMV-specific CD8 T cells during CHI ( Figure S2C , D ) . The MFI and frequency of PD-1 on HIV-specific T cells were significantly lower in ECs ( 54 . 2%; MFI = 3479 ) compared to AHI ( 69 . 8% , P = 0 . 0004; MFI = 5742 , P<0 . 0042 ) , CHI ( 86 . 0% , P<0 . 0001; MFI = 8377 , P = 0 . 0002 ) and ST patients ( 76 . 1% , P = 0 . 03; MFI = 5097 , P = 0 . 006 ) . Total CD8 T cells in CHI subjects and ECs also included a significantly higher proportion of cells expressing high levels of CD160 compared to levels observed in uninfected controls ( MFI = 1251 and 24 . 4% in uninfected individuals; MFI = 1340 and25 . 3% in AHI; MFI = 1983and 35 . 4% in CHI; MFI = 1788 and 36 . 0% in ECs ) ( Figure S3 ) . As shown for antigen-specific responses , total CD8 T cells upregulated CD160 expression only during the chronic stage of infection . CD160 expression was restricted to terminally differentiated ( CD45RA+CD27−CCR7− ) and memory CD8 T cells ( CD45RA−CD27+/−CCR7+/− ) whereas naïve CD8 T cells ( CD45RA+CD27+CCR7+ ) did not express this protein ( Figure S4 ) . Of note , while PD-1 levels were upregulated on HIV-specific and total CD8 T cells in AHI and CHI compared to uninfected subjects , elevated CD160 levels were limited to CHI . Our results highlight a difference in the timing of expression of these molecules during the natural history of HIV infection . We measured the co-expression of CD160 and PD-1 on HIV and CMV-specific CD8 T cells in the groups of subjects described above . Four distinct cell subsets were identified: CD160−PD-1− ( DN ) , CD160−PD-1+ ( SP-PD-1 ) , CD160+PD-1− ( SP-CD160 ) and CD160+PD-1+ ( DP ) ( Figure 1A , Figure S5A , B ) . Frequencies of DN and DP subsets within HIV-specific T cells showed significant differences when comparing cross-sectionally HIV-infected subjects at different stages of disease ( AHI vs CHI ) . The percentages of HIV-specific DN cells were highest in AHI ( 20 . 2% range , 4 . 7–31 . 3 ) and lowest in CHI ( 3 . 2% range , 0 . 7–19 . 1 ) ( P = 0 . 0002 ) ( Figure 1B; upper left panel ) . In contrast , frequencies of HIV-specific DP cells were highest in CHI ( 59 . 5% range , 21 . 5–78 . 2 ) and their numbers were the lowest in AHI ( 24 . 8% range , 9 . 4–37 . 0 ) ( P = 0 . 0001 ) ( Figure 1B; upper right panel ) . In this cross-sectional analysis , the frequencies of DN and DP subsets within CMV-specific T cells did not vary between acute and chronic infection ( Figure 1B ) . Elite controllers also showed high frequencies of HIV-specific T cells with a DP phenotype; however frequencies were significantly lower compared to CHI ( P = 0 . 004 ) ( Figure 1B; upper right panel ) . As for the two SP subsets , their frequencies significantly shifted when comparing individuals at different stages of disease . The frequency of SP-PD-1 HIV-specific CD8 T cells was highest during AHI ( 47 . 1% ) and significantly decreased in CHI ( 25 . 5%; P = 0 . 0001 ) and ECs ( 14 . 0% , P = 0 . 008 ) while the frequencies of SP-CD160 CD8 T cells were higher in ECs compared to other study groups ( Figure 1B; lower panels ) . The frequencies of SP-PD-1 and DP subsets from ST subjects did not significantly differ from those observed in CHI and EC . Interestingly , the frequency of SP-PD-1 HIV-specific CD8 T cells did not differ from that of SP-PD-1 T cells recognizing CMV epitopes at all stages of HIV infection ( Figure 1B; lower right panel ) . These results indicated that HIV-specific CD8 T cells expressing PD-1 encompass multiple subsets . Importantly , the phenotype of these cells was shown to be different when comparing CD8 T cells in acute and chronic HIV infection . Moreover , the level of PD-1 expression was highest on CD160+CD8 T cells ( Figure S5C ) , further reinforcing the notion that DP cells are at an advanced stage of exhaustion and raising the possibility that SP-PD-1 cells encompass recently activated T cells [32]–[33] The lower frequencies of HIV-specific DP CD8 T cells in ECs compared to CHI subjects ( Figure 1 ) suggested the involvement of CD160 and PD-1 co-expression in antigen-specific T cell dysfunction as previously shown in mice models of chronic viral infection [20]–[21] and suggested recently in HIV-infected subjects [27] , [30] . A longitudinal analysis was performed to confirm the dynamic evolution of these phenotypes during different disease stages . We assessed the frequencies of CMV , EBV and HIV-specific CD8 T cells expressing CD160 and/or PD-1 in 5 HIV-infected subjects during AHI ( <3 months on infection ) and CHI ( >6 months of infection ) ( Figure 2A ) . The frequency of HIV-specific PD-1 expressing CD8 T cells ( SP-PD-1 and DP ) remained stable over time ( Figure S6 ) . However , when we assessed the two subsets of PD-1+ cells , we observed a significant decrease in the frequencies of SP-PD-1 HIV-specific CD8 T cells ( P = 0 . 016 ) over the course of infection whereas the frequencies of DP CD8 T cells significantly increased ( P = 0 . 016 ) from AHI to CHI ( Figure 2B; right panels ) confirming results obtained in the cross-sectional analysis . In contrast , the proportion of CD160+ HIV-specific CD8 T cells ( SP-CD160 and DP ) increased with disease progression ( Figure S6 ) . Percentages of DN and SP-CD160 HIV-specific CD8 T cells increased from AHI to CHI ( P = 0 . 016 ) ( Figure 2B; left panels ) . The distribution of CD160 and PD-1 subsets on cells specific for CMV and EBV epitopes did not significantly change from AHI to CHI ( Figure 2C ) . Taken together , these results showed a dynamic evolution of the frequency and distribution of CD160 and/or PD-1 expression on HIV-specific CD8 T cells during infection . Both cross-sectional and longitudinal results confirmed that the distribution of CD160 and PD-1 expressing subsets was predominantly SP-PD-1 in AHI and DP in chronic HIV infection . Our results strongly indicated that the simultaneous expression of CD160 and PD-1 could constitute a marker of T cell exhaustion and disease progression . A recent study published by Youngblood et al . [34] reinforced this observation by showing that persistent TCR signalling results in sustained PD-1 expression by maintaining PD-1regulatory regions accessible to transcription factors . We next assessed the effector function of these different subsets by intracellular cytokine staining ( ICS ) for IFNγ and TNFαsecretion and measured the expression of CD107a following stimulation with SEB , CMVpp65 and HIV peptides in viremic HIV-infected subjects . Figure 3A depicts the percentage of cytokine secreting cells within the total CD8 T cell population upon SEB stimulation . Frequencies of CD8 T cells producing TNFα , IFNγ or upregulating CD107a were significantly higher in DN , SP-PD-1 and SP-CD160 compared to frequencies observed in the DP subset ( P<0 . 05 and P<0 . 005 , #or ## represents significant change in cytokine production compared to DP ) . Furthermore cells with a DN phenotype had higher frequencies of functional CD8 T cells when compared to SP-CD160 and SP-PD-1 following stimulation with SEB . In summary , the expression of either CD160 or PD-1 on CD8 T cells identified a T cell subset having lower levels of effector function ( IFNγ , TNFα ) or degranulation ( CD107a ) . Importantly , our results indicated that cells expressing both CD160 and PD-1 were more dysfunctional than cells expressing either one of these two molecules . The same hierarchy of functionality was observed when comparing the four different subsets of T cells following stimulation with individual CMV peptides ( Figure 3B ) . As observed above , DN cells showed the highest frequencies of cells that produced TNFα ( 19 . 2% ) , IFNγ ( 11 . 5% ) , and upregulated CD107a ( 17 . 6% ) when compared to DP cells ( TNFα0 . 4% , IFNγ0 . 5% , CD107a 0 . 5% ) . SP-PD-1 ( TNFα = 1 . 7% , IFNγ = 2 . 9% , CD107a = 2 . 5% ) and SP-CD160 ( TNFα = 1 . 3% , IFNγ = 1 . 5% , CD107a = 3 . 8% ) also showed lower frequencies of cytokine producing cells than DN and importantly higher frequencies than DP cells . Triggering of HIV-specific cells yielded similar results to SEB and CMV stimulated T cells . As noted above , DP ( TNFα: 0 . 8% , IFNγ:1 . 2% , CD107a:2 . 6% ) and DN ( TNFα:3 . 5% , IFNγ:5 . 4% , CD107a:13 . 4% ) cells showed the most significant difference in the frequency of cytokine producing cells with a downregulation of TNFα , IFNγand CD107a expression ( Figure 3C ) . Following stimulation with HIV peptides , the SP-PD-1 ( TNFα:1 . 9% , IFNγ:5 . 4% , CD107a:7 . 0% ) subset included significantly higher frequencies of IFNγ , TNFα secreting cells as well as CD107a+ degranulating cells as compared with the DP subset ( Figure 3C; ##represents significant change in cytokine production compared to DP ) . As observed for SEB and CMV peptide stimulated CD8 T cells , DP CD8 T cells were consistently the least functional subset . Together these results showed that antigen-specific CD8 T cells with a DP phenotype were less functional than SP-PD-1 expressing CD8 T cells as shown by their lower responses to all three T cell activation signals . These results provide evidence that co-expression of CD160 and PD-1identified dysfunctional CD8 T cells . This degree of dysfunctionality progressively increases with the co-expression of additional immune inhibitory markers on CMV and HIV-specific T cells . HVEM is the natural ligand of CD160 [35] . We therefore performed experiments aimed at determining whether interfering with CD160 engagement by HVEM allowed dysfunctional T cells to recover their effector T cell function . PBMCs were stimulated with HLA-restricted CMV and HIV optimal peptides in a 6-day CFSE assay in the presence or absence of αHVEM together with or without αPD-L1 blocking antibodies ( Figure 4A ) . We measured the expression of HVEM on monocytes , mDCs and pDCs and observed a significant upregulation of HVEM surface expression on monocytes and pDCs from CHI individuals compared to healthy controls . Similar findings were observed when measuring the expression of PD-L1 on monocytes ( P<0 . 05 ) ( Figure S7 ) . Our results showed that blocking CD160 interaction with HVEM significantly enhanced CMV and HIV-specific CD8 T cell proliferation ( Figure 4B , C ) . We observed a median fold increase of 10 . 1 ( P<0 . 0001 ) and 4 . 9 ( P<0 . 0001 ) in CMV and HIV-specific CD8 T cell proliferation respectively , compared to isotype controls . Blocking the PD-1/PD-L1 pathway led to a statistically significant enhancement of HIV-specific CD8 T cell proliferation by a factor of 1 . 2 ( P = 0 . 02 ) ; hence CD160/HVEM blockade was more potent than PD-1/PD-L1 blockade at restoring HIV-specific CD8 T cell proliferation . PBMCs cultured with both blocking antibodies ( αPD-L1 and αHVEM ) also significantly enhanced T cell proliferation however , the effect was not synergistic compared to using αHVEM alone ( P<0 . 0001 ) ( Figure 4B , C ) . We analyzed the coexpression of BTLA and CD160 on HIV-specific CD8 T cells during chronic HIV infection and found that the frequency of CD160 ( 35 . 5% ) was significantly greater than BTLA ( 3 . 9%; P<0 . 0001 ) suggesting that using αHVEM preferentially disrupts the CD160/HVEM axis ( Figure S8 ) . Controls showed that the αHVEM blocking antibody did not induce T cell activation in the absence of the T cell cognate peptide . Supernatants harvested following the 6-day CFSE assay were used to assess cytokine production in the presence or absence of PD-1 and or CD160 engagement by their respective ligands . We found that levels of IFNγ , IL-4 and IL-10 were significantly increased compared to isotype controls in conditions where αHVEM was added to the T cell cultures ( IFNγ , P = 0 . 001; IL-4 , P = 0 . 03; IL-10 , P = 0 . 003 ) ( Figure S9 ) . Although TNFα production was increased in conditions where αHVEM was present , the levels were not statistically different compared with the isotype control ( TNFα , P = 0 . 054 ) . The levels of IL-2 production did not significantly increase upon antigen-specific stimulation in the presence of HVEM blockade most likely due to the consumption of this cytokine by proliferating T cells . These results confirmed that CD160 was implicated in HIV-specific CD8 T cell exhaustion . Blocking its interaction with HVEM restored the proliferation and cytokine production of antigen-specific CD8 T cells . Gene array profiling was performed on sorted CD8 T cell subsets based on CD160 and PD-1 co-expression in 4 HIV viremic individuals to determine if the functional defects observed in DP cells were the consequence of a distinct gene expression signature that was associated with signal transduction pathways that regulate T cell survival , turnover and function . Unsupervised cluster analysis showed that both DP and SP-PD-1 subsets clustered apart to create two statistically significant populations with unique transcriptional profiles ( Figure 5A ) . The heatmap lists the top 39 genes expressed at significantly different levels between both subsets ( P<0 . 05 ) . The results showed that genes upregulated in DP cells include those involved in the inhibition of several survival pathways . Importantly , SUMO2 ( Small Ubiquitin-like modifier ) was upregulated in DP CD8 T cells compared to SP-PD-1 . This enzyme upregulates the activity of PIAS ( protein inhibitor of activated STAT ) molecules which are responsible for the inhibition of STATs ( Signal Transducer and Activator of Transcription ) including STAT5 , a molecule directly downstream of γ chain cytokine receptors such as IL-7 and IL-15 [36]–[38] . The inhibition of the STAT5 pathway was confirmed by the downregulation of bcl-2 in DP cells ( Figure 5B ) . Moreover DP cells upregulated the expression of KIF7 , an antagonist of hedgehog the positive regulator of Wnt signaling [39] and several cell surface negative regulatory molecules including KIR2DL3 known to express ITIM motifs in the cytoplasmic tail [40] . In contrast , SP-PD-1 CD8 T cells upregulated the expression of several T cell activation markers including HAVCR2 ( Tim-3 ) , CTLA-4 , LAT , CCR1 and TNFRS25 [26]–[27] , [41]–[42] . In addition , positive regulators of Wnt/Notch signalling including Wnt7A and AXIN2 were upregulated in SP-PD-1 compared to DP CD8 T cells [43] . A network analysis of differentially expressed genes between DP and SP-PD-1 subsets showed a significant downregulation of signal transduction pathways enriched in genes that regulate T cell survival ( IL-15 , IL-7R , PIM3 , bcl-2 , all part of the STAT-5A pathway ) and T cell effector function ( LTβ , IL-18RAP , IL-18R1 , CXCL16 ) in the DP subset ( Figure 5B ) . Interestingly , the expression of NFκB was downregulated in DP compared to SP-PD-1 further confirming the advanced state of exhaustion and the unique identity of this novel subset in chronic HIV infection . Of note , TNFα production , which triggers NFκB [44] , was downregulated in CD160+ CD8 T cells . Moreover , this network analysis confirmed the upregulation of several molecules with inhibitory functions in the DP subset such as the inhibitory KIR family ( KIR2DL1 , 2DL2 , 2DL3 , 2DL4 , KIR3DL1 , 3DL3 ) [40] , 2B4 as well as members of the KLR family of proteins which are all associated to senescence . We confirmed by flow cytometry the increased expression of inhibitory KIR2DL2/KIR2DL3 in the DP compared to SP-PD-1 subset ( p = 0 . 038 ) ( Figure 5C ) . Taken together , gene expression analysis of CD160+PD-1+ CD8 T cells showed a gene expression signature that comprises several inhibitors of survival signal transduction pathways ( STAT-5 and Wnt/Notch ) and the increased expression of multiple immune inhibitory molecules . In contrast , SP-PD-1 CD8 T cells showed a transcriptional profile reminiscent of recently activated T cells .
The results presented here show that CD160 was upregulated on CD8 T cells during HIV infection . We identified 4 distinct subsets of CD8 T cells: DN , SP-PD-1 , SP-CD160 and DP . Importantly , only CD8 lymphocytes that co-expressed CD160 and PD-1 had functional features and transcriptional profiles of exhausted cells . Recent studies have described an accumulation of inhibitory molecules on HIV-specific CD8 T cells . However , we show here that the distribution of cells expressing one or more of these molecules significantly shifted during the course of HIV infection . We show that SP-PD-1 cells increased in numbers in AHI while cells co-expressing CD160 and PD-1 were the dominant cell subset in CHI . This increased frequency of DP cells was associated with HIV disease progression and T cell dysfunction . Cells within the CD160+PD-1+ subset were less functional than SP-PD-1 and SP-CD160 as shown by the reduced frequency of cytokine secretion upon TCR triggering . Our gene array data confirmed the unique exhausted phenotype of the DP subset as these cells expressed transcriptional programs that were highlighted by the downregulation of the NFκB transcriptional node , strongly associated to T cell survival , and the upregulation of cell surface inhibitory KIR expression . This allowed us for the first time to provide molecular evidences for differences that demarcate cells expressing these inhibitory molecules as a consequence of T cell activation or those that express these molecules when they are functionally exhausted . We also show that CMV and HIV-specific CD8 T cell proliferation and cytokine secretion were rescued after blocking the engagement of CD160 with its natural ligand ( HVEM ) . These results confirmed that functional exhaustion of T cells results from the progressive accumulation of several molecules that negatively impact on T cell activation . The temporal accumulation of these negative regulators results from chronic exposure to HIV and other molecules that trigger hyper-immune activation [45] since CD160 expression on T cells is observed mostly during the chronic phase of infection Along with LIGHT , LTα , HSVgD , and BTLA , CD160 is also a ligand of HVEM . The interaction of LIGHT with HVEM delivers a co-stimulatory signal by triggering NFκB whereas the binding of CD160 or BTLA with this ligand delivers an inhibitory signal to CD4 T cells [35] , [46] , most probably by competing with LIGHT [47] for binding to HVEM . Although conflicting results regarding the function of CD160 have been reported [48]–[49] , [50]–[51] , recent findings are consistent with an inhibitory function for this molecule when expressed on T cells [20]–[21] , [35] . The fact that SP-CD160 and SP-PD-1 subsets are expressed at comparable levels on HIV and CMV-specific T cells suggest that these cells are still functionally competent . Future work will compare the phenotype of HIV-specific CD8 T cells to other acute and chronic viral infections with the aim of understanding whether the observed phenotypic distribution is unique to HIV-specific CD8 T cells . Moreover since SP-PD-1 can still mount polyfunctional responses upon TCR triggering with cognate antigen , and the observation that HIV-specific T cells in ECs exhibit mostly an SP-CD160 phenotype further confirms the functionality of these subsets . DP cells express the highest levels of PD-1 when compared to SP-PD-1 . DP CD8 T cells are hence a unique dysfunctional T cell subset , as highlighted by the downregulation of several transcriptional nodes ( STAT5 , Notch-Wnt , NFκB ) that regulate T cell survival and effector function as confirmed by flow cytometry and T cell functional assays . Differences in the functionality of DP and SP-PD-1 subsets could not be accounted for by their distribution in different memory or effector T cell compartments . Indeed our results ( Figure S4 ) showed that DP and SP-PD-1 cells were found in TTM , TEM and late differentiated T cells all known to be endowed mostly with effector functions thereby confirming the data by Yamamoto et al . [30] . These results are consistent with those generated in the LCMV model whereby the frequency of CD160+PD-1+CD8 T cells increases during CHI leading to an accumulation of dysfunctional HIV-specific CD8 T cells [20]–[21] . The expression of PD-1 in ST patients is significantly higher than that observed in ECs highlighting the ongoing viral replication in tissues ( Figure S2 ) . The higher levels of PD-1 on HIV-specific CD8 T cells from ST subjects most probably contribute to the dysfunction of these cells while DP cells from EC subjects still remain functional [17] . Previous studies have shown that CD160 and PD-1 interact with molecules downstream of the TCR [35] , [52] . Following the activation of CD4 T cells with αCD3/CD28 , ligation of CD160 with HVEM reduced the phosphorylation of tyrosine residues on several substrates such as CD3ζ . This decreased expression and phosphorylation of CD3ζ has been associated to T cell anergy and dysfunction [53]–[54] . Our results confirm that the presence of both CD160 and PD-1 on the surface of cells is required for the inhibition of TCR mediated signalling . In that context , we observed higher frequencies of functional antigen-specific CD8 T cells in lymphocytes negative for both CD160 and PD-1 , whereas subsets that expressed either molecule alone were significantly less functional . It is important to note that both CD160 and PD-1 are upregulated upon T cell activation [55] . Hence it is more than likely that SP-PD-1 and SP-CD160 cells correspond to recently activated T cells that have upregulated those inhibitory receptors ( PD-1 , CD160 ) to control T cell activation as part of a homeostatic T cell response . Blocking the interaction of CD160 and HVEM significantly enhanced CMV and HIV-specific proliferation and cytokine production further reinforcing the notion that CD160 acts as a negative regulator of T cell function . The magnitude of increase in CMV responses upon addition of αHVEM was more important than HIV-specific CD8 T cell responses ( Figure 4 ) . This is most probably due to the higher frequencies of DN and SP-CD160 cells within the CMV-specific T cell pool as compared to HIV-specific T cells . In addition , HIV-specific T cells include higher frequencies of DP cells compared to CMV-specific CD8 T cells . We found that the increase in proliferation that resulted from blocking CD160 engagement with HVEM was greater than that observed upon blocking PD-1 engagement with PD-L1 [24] , [30] . As we have shown that CD160 is expressed at much higher levels than BTLA ( the other ligand of HVEM ) , it is most likely that the rescue of HIV-specific CD8 T cell responses observed after addition of αHVEM targets mostly the interaction of HVEM with CD160 . As previously shown , PD-1 is mostly expressed on DP cells during CHI . In CHI , the frequency of DP HIV-specific CD8 T cells is significantly higher than DP CMV-specific CD8 T cells ( p<0 . 0001 ) . HIV-specific DP cells are at an advanced stage of exhaustion and simultaneously express other negative regulatory molecules ( elevated expression of KIR receptors ) , which could also contribute to T cell dysfunction . These cells , as shown from our transcriptional profiling , are hence truly exhausted and are at an irreversible stage of T cell dysfunction . Several studies have observed that restoration of HIV-specific CD8 T cell proliferation and cytokine secretion of T cells specific for different epitopes [23]–[25] was variable . Our results suggest that the presence of multiple subsets of T cells expressing PD-1 with other negative regulators could account for this variability . For instance , in the murine LCMV infection model and during chronic HCV infection , subsets of PD-1hi and PD-1intexpressing cells have been identified . Blocking experiments have shown that only CD8 T cells that expressed intermediate levels of PD-1 were responsive to PD-1/PD-L1 blockade suggesting that not all specificities reverted towards a functional phenotype [56]–[57] . As noted above , PD-1 and CD160 are upregulated upon T cell activation . Transcriptional profiling helped elucidate the differences between cells that upregulate PD-1 as a result of T cell activation and PD-1hi exhausted T cells . Indeed SP-PD-1 cells also expressed several other T cell activation markers ( CTLA-4 , Tim-3 , CCR1 , TNFRSF25 ) as well as other molecules associated with T cell survival ( AXIN2 , Wnt7A ) . In contrast , the KIR family of cell surface receptors clearly demarcated PD-1hi ( DP cells ) exhausted cells from SP-PD-1 activated T cells [58] . Interestingly , only KIR genes with ITIM motifs ( KIR2DL1 , 2DL3 , 2DL4 , 3DL1 , 3DL2 , and 3DL3 ) were found upregulated on CD8 T cells which co-expressed CD160 and PD-1 . The DP phenotype was associated with the down regulation of the NFκB survival pathway . As noted above , NFκB activity is triggered by LIGHT that competes with CD160 for binding to HVEM . These results support the view that DP lymphocytes express a large array of immune inhibitors and prevent CD8 T cells from acquiring a fully functional state . Our gene array results further confirmed the significant differences that demarcate DP from SP-PD-1 CD8 T cells . The former represent exhausted T cells while the latter are characterized by the expression of T cell activation markers and evidence for the induction of several pathways of T cell activation . Our findings demonstrate that T cell exhaustion during chronic viral diseases results from the progressive temporal accumulation of multiple negative regulators of T cell activation and their interaction with their ligands . Understanding the contribution of these multiple inhibitory signals will be essential to properly define exhaustion and to determine whether these pathways converge to inhibit T cell activation by targeting multiple cellular pathways . In that context , system biology and transcriptional profiling have provided essential tools to dissect the functional status of cells expressing the different negative regulators of T cell activation . Functional restoration of exhausted T cell subsets will require combination therapies that target distinct sets of receptors at different stages of infection .
Written informed consent was provided by study participants and approved by the University of Montreal Health Center ethics review board ( CRCHUM ) . Research conformed to ethical guidelines established by the ethics committee of the University of Montreal Health Center . The study population included 38 HIV-1 subtype B infected individuals at various stages of infection and 7 HIV-uninfected donors ( Table 1 and 2 ) . HIV-infected patients were categorized into 4 subgroups: Elite controllers ( ECs; n = 10 ) infected for more than 7 years with undetectable viremia , successfully treated ( ST; n = 12 ) and aviremic individuals , subjects with acute HIV infection ( AHI; n = 7 ) analyzed within 3 months of infection [59] and 9 chronically progressing subjects ( CHI ) infected for more than 6 months based on CD4 T cell counts under 500/mm3or declining CD4 T cell counts . All HIV infected subjects with the exception of ST , were naïve to antiretroviral therapy at the time of testing . Plasma viral loads were measured with the Amplicor HIV-1 Monitor Ultrasensitive method with a limit of detection of 50 HIV-1 RNA copies/mL of plasma ( Roche Diagnostics , Mississauga , Canada ) . Absolute CD4 counts were quantified by the BD Multitest ( CD3/CD4/CD8/CD45RA ) using a FACSCanto ( BD ) . Soluble pMHC monomers were generated as previously described ( Montreal , Canada ) [60] . The peptides and tetramers used to analyze the CMV , EBV and HIV-specific CD8 T cell responses were: NLVPMVATV ( A*02 CMV ) , TPRVTGGGAM ( B*07 CMV ) , GLCTLVAML ( A*02 EBV ) , RAKFKQLL ( B*08 EBV ) , FLGKIWPSYK ( A*02 Gag ) , ILKEPVHGV ( A*02 Pol ) , SLYNTVATL ( A*02 Gag ) , RLRPGGKKK ( A*03 Gag ) , AIFQSSMTK ( A*03 RT ) , QVPLRPMTYK ( A*03 NEF ) , RYPLTFGWCF ( A*23 NEF ) , RPGGKKKYKL ( B*07 Gag ) , TPGPGVRYPL ( B*07 NEF ) , SPAIFQSSM ( B*07 RT ) , GEIYKRWII ( B*08 Gag ) , FLKEKGGL ( B*08 NEF ) , DCKTILKAL ( B*08 Gag ) , RRWIQLGLQK ( B*27 Gag ) and YPLTFGWCF ( B*35 NEF ) . PBMCs were resuspended in PBS containing 2% FCS and stained with Tetramer-PE at 0 . 3 µg per 106 cells . The following cocktail was used for phenotyping CD8 T cells: αCD160FITC ( BD ) , αCD3Alexa 700 ( BD ) , αCD8 PB ( BD ) , αCD45RA ECD ( Beckman Coulter ) , αCD27 APC-Cy7 ( eBioscience ) , αCCR7 PE-Cy7 ( BD ) , αCD158b PE ( KIR2DL2 , KIR2DL3 ) ( BD ) andαPD-1 APC ( eBioscience ) . Monocytes , mDCs and pDCs were labelled using αCD3 , αCD16 and αCD19 Alexa700 ( BD ) , αCD14 PB ( BD ) , αHLA-DR APC-Cy7 , αCD11c APC , αCD123 PE ( BD ) , αPD-L1 PE-Cy7 ( BD ) and αHVEM FITC ( MBL ) . Dead cells were eliminated with an amine-reactive viability dye ( LIVE/DEAD , Invitrogen ) . We acquired a minimum of 1×106 events for all cytometry-related experiments using a BD LSRII flow cytometer and analyzed the results with FlowJo 9 . 1 ( Treestar ) . Optimal peptides used for ICS and CFSE assays were identical to the ones folded in the pMHC monomers . PBMCs were stimulated with 5 ug/ml of CMV and HIV-specific peptides as described previously [24] . The cocktail used for ICS was: Tetramer PE , αCD160 FITC ( BD ) , αTNFα Alexa 700 ( BD ) , αIFN-γ PE-Cy7 ( BD ) , αPD-1 APC , αCD3 PB and αCD8 ECD ( Caltag ) . For CFSE , we stimulated PBMCs with CMV and HIV peptides for 6 days in the presence of 10 µg/ml of αPD-L1 ( eBioscience ) , αHVEM ( R&D systems ) and polyclonal goat or monoclonal mouse IgG1 isotype controls ( R&D systems ) . As described in the manufacturer's protocol , we used a cytokine bead array ( CBA ) ( BD ) assay to measure the concentrations of IL-4 , IL-2 , IL-10 , TNFα and IFNγ in the supernatants harvested at the end of the CFSE assay . CD8 T cells subsets expressing CD160 and/or PD-1 were sorted from 4 HIV chronically infected individuals using BD FACS ARIA , lysed in RLT buffer and stored at −80°C . Total RNA was purified using RNA extraction kits ( RNeasy Micro Kit , Qiagen ) . Quantification was performed using a spectrophotometer ( NanoDrop Technologies ) and RNA quality was assessed using the Experion automated electrophoresis system ( Bio-Rad ) . Total RNA was amplified using the Illumina TotalPrep RNA Amplification kit [61] . Biotinylated cRNA was hybridized onto Illumina Human RefSeq-8 BeadChips and quantified using Illumina BeadStation 500GX scanner and Illumina BeadScan software . Gene expression data was analyzed using Bioconductor ( www . bioconductor . org ) [62] . The R software package was used for pre-processing to filter out genes with intensities below background in all samples , minimum-replace ( a surrogate-replacement policy ) values below background using the mean background value of the built-in Illumina probe controls as an alternative to background subtraction , reduce “over inflated” expression ratios and finally quantile-normalize the gene intensities . Out of the 24526 initial probe set , 9070 probes were left after the filtering steps . The resulting matrix was log2 transformed and used as input for linear modeling using Bioconductor's limma package which estimates the fold-change between predefined groups by fitting a linear model and using an empirical Bayes method to moderate standard errors of the estimated log-fold changes for expression values from each gene [63]–[64] . P values from the resulting comparison were adjusted for multiple testing according to the method of Benjamini and Hochberg . Gene networks were generated using Ingenuity Pathway Analysis ( www . ingenuity . com ) . A dataset containing gene identifiers and corresponding statistical values were uploaded to the application . Each gene identifier was mapped to its corresponding gene in the Ingenuity Pathways Knowledge Base . Genes obtained from this analysis were overlaid onto a global molecular network . Networks of these focused genes were then algorithmically generated based on their connectivity . Statistical analysis and graphical presentation was performed using GraphPad Prism 5 . 0c ( GraphPad software , San Diego , CA ) FlowJo 9 . 1 ( Treestar ) and SPICE 5 . 1 ( http://exon . niaid . nih . gov ) [65] . Two-tailed Unpaired tor Mann-Whitney tests were used to assess between-group differences ( Figure 1 ) . Two-tailed Wilcoxon matched pairs test was used to assess differences in the relative frequency of subsets over time ( Figure 2 ) , in the functionality between CD160 and PD-1 expressing subsets ( Figure 3 ) and to assess differences in proliferative responses following co-culture with blocking antibodies ( Figure 4 ) . To determine if the variables analyzed came from a Gaussian distribution we applied the D'Agostino and Pearson's normality test . P-values less than 0 . 05 were considered statistically significant . | HIV infection is widely known to cause generalized immune activation and immune exhaustion ultimately leading to HIV disease progression . Several studies have suggested over the years that the accumulation of inhibitory signalling proteins on the surface of responding cells is linked to immune exhaustion in HIV . It has become paramount to distinguish functionally exhausted CD8 T cells from activated HIV-specific CD8 T cells because both cell types have different fates . Using specific cell surface markers , we were able to identify these different cell types and show that HIV-infected patients accumulate dysfunctional CD8 T cells over time . Importantly , we show that this dysfunction is reversible . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"medicine",
"immune",
"cells",
"immune",
"activation",
"retrovirology",
"and",
"hiv",
"immunopathogenesis",
"immunity",
"to",
"infections",
"immunology",
"adaptive",
"immunity",
"sexually",
"transmitted",
"diseases",
"immunomodulation",
"infectious",
"diseases",
"aids",
"hiv",
"t",
"cells",
"biology",
"immune",
"response",
"clinical",
"immunology",
"immunity",
"immune",
"deficiency",
"viral",
"diseases"
] | 2012 | CD160 and PD-1 Co-Expression on HIV-Specific CD8 T Cells Defines a Subset with Advanced Dysfunction |
Phenotypic variation within an isogenic bacterial population is thought to ensure the survival of a subset of cells in adverse conditions . The opportunistic pathogen Pseudomonas aeruginosa variably expresses several phenotypes , including antibiotic resistance , biofilm formation , and the production of CupA fimbriae . Here we describe a previously unidentified bistable switch in P . aeruginosa . This switch controls the expression of a diverse set of genes , including aprA , which encodes the secreted virulence factor alkaline protease . We present evidence that bistable expression of PA2432 , herein named bexR ( bistable expression regulator ) , which encodes a LysR-type transcription regulator , controls this switch . In particular , using DNA microarrays , quantitative RT–PCR analysis , chromatin immunoprecipitation , and reporter gene fusions , we identify genes directly under the control of BexR and show that these genes are bistably expressed . Furthermore , we show that bexR is itself bistably expressed and positively autoregulated . Finally , using single-cell analyses of a GFP reporter fusion , we present evidence that positive autoregulation of bexR is necessary for bistable expression of the BexR regulon . Our findings suggest that a positive feedback loop involving a LysR-type transcription regulator serves as the basis for an epigenetic switch that controls virulence gene expression in P . aeruginosa .
The Gram-negative bacterium Pseudomonas aeruginosa is an opportunistic pathogen of humans . It can cause infection in a wide variety of tissues in the immunocompromised host , and is the leading cause of morbidity and mortality in cystic fibrosis ( CF ) patients [1] . This breadth of infectious capacity is thought to result from differential gene expression , as genomic variability between clinical and environmental isolates is low and the genome of P . aeruginosa encodes a high proportion of transcription regulators [2] , [3] . Studying the mechanisms and outcomes of transcription regulation in P . aeruginosa may offer some insight into how cohorts of virulence factors are coordinately expressed to influence pathogenesis in a range of pseudomonal infections . Bacteria are traditionally thought to use transcription regulation to adapt to changing environmental conditions , such as the presence of a new carbon or energy source , a change in temperature or pH , or introduction to a host environment . However , in harsh environmental conditions that exert a sudden selective pressure on a population of cells , the time needed to respond using a genetic regulatory network may prove fatal . The ability of isogenic populations of bacteria to exhibit phenotypic variation allows them to cope with such situations by pre-adapting a subset of the population to the sudden introduction of harsh conditions . Several examples of phenotypic variation in P . aeruginosa have been identified , such as the phase-variable expression of the cupA fimbrial gene cluster under anaerobic conditions and the transient formation of antibiotic resistant , hyperadherent rough small-colony variants under antibiotic exposure [4]–[7] . These phenotypes may contribute to the ability of infecting bacteria to withstand chemical or mechanical insults encountered during colonization of the CF lung . Examples such as these suggest that phenotypic variation by P . aeruginosa allows the organism to thrive in a complex environment . However , the mechanisms by which these phenotypes are variably expressed are unknown . Phenotypic variation in bacteria can arise from a variety of mechanisms , both genetic and epigenetic in nature . Classical phase-variation is thought to be genetically mediated , such as the variable expression of the flagellum in Salmonella enterica serovar Typhimurium , which is mediated by specifically catalyzed changes in promoter DNA orientation [8] . Phase-variation can also be mediated by epigenetic mechanisms , such as the one involving DNA methylation that controls the phase-variable expression of pyelonephritis-associated pili genes in uropathogenic Escherichia coli [9] , [10] . Phenotypic heterogeneity can arise in the absence of DNA sequence variation or DNA modification in bistable systems ( i . e . systems that can exist in one of two alternative expression states , and reversibly switch between them ) , such as in the case of the lysogenic switch of bacteriophage λ [11] , [12] . Bistability can arise when there exists a mechanism for amplifying differences in protein levels between individual cells and stably propagating these differences to daughter cells ( reviewed in [13] ) . The bistable expression of genes can be achieved using a positive regulatory feedback loop , as is the case in the development of competence under nutrient limitation in Bacillus subtilis; positive feedback of ComK , the master regulator of competence , is required for bistable development of competence in B . subtilis [14] , [15] . Thus , the architecture of a particular gene regulatory circuit can enable stochastic , reversible differentiation of subsets of bacterial populations into distinct cell types . Here we uncover a previously unidentified bistable switch in P . aeruginosa controlled by BexR , a LysR-type transcription regulator . We demonstrate that bexR is itself bistably expressed in a BexR-dependent manner and that BexR positively regulates the expression of its own gene . Using DNA microarrays and quantitative real-time RT-PCR ( qRT-PCR ) , we define the bistable regulon of BexR , which contains a diverse set of genes and includes aprA , which encodes the virulence factor alkaline protease . We show further that BexR acts directly at the promoters of many of its regulatory targets , including that of its own gene . Finally , we describe a series of single-cell population analyses that suggest that this bistable switch requires bexR autoregulation . We propose a model for the BexR switch in which positive feedback amplifies bexR expression in a stochastically determined subset of cells , giving rise to bistable expression of BexR target genes in an isogenic population .
In the course of unrelated microarray experiments , we observed a small set of genes that exhibited variable expression between replicates of wild-type cultures of P . aeruginosa strain PAO1 ( data not shown ) . This set includes PA1202 , which encodes a hypothetical protein with homology to a predicted hydrolase of Escherichia coli , and PA2432 ( herein named bexR for bistable expression regulator ) , which is predicted to encode a member of the LysR family of transcription regulators . To confirm that PA1202 is expressed in a variable manner , we constructed a strain of PAO1 in which lacZ was placed downstream of the PA1202 gene ( Figure 1A ) . This strain exhibits reversible bistable expression of the lacZ reporter . Specifically , wild-type cells of this reporter strain give rise to both blue ( “ON” ) and white ( “OFF” ) colonies on LB agar plates containing X-Gal ( Figure 1B ) . When re-streaked on LB agar with X-Gal , ON colonies give rise to both ON and OFF colonies , and OFF colonies give rise to both OFF and ON colonies . Because our initial microarray analyses suggested that bexR , which encodes a putative transcription activator , co-varied with PA1202 , we hypothesized that BexR may positively regulate expression of PA1202 and that bistable expression of bexR may be responsible for the observed bistability in PA1202 expression . To begin to test this hypothesis , we constructed an unmarked in-frame deletion of bexR in PAO1 PA1202 lacZ . Compared to the wild-type reporter strain , the ΔbexR mutant exhibits constitutively low-level expression of PA1202 ( Figure 1B ) . Ectopic expression of bexR in the ΔbexR mutant resulted in increased PA1202 expression ( Figure 1C ) , suggesting that BexR positively regulates expression of PA1202 . However , bistable expression of PA1202 is lost when bexR is expressed ectopically; PAO1 ΔbexR PA1202 lacZ grows only as ON colonies on LB agar with X-Gal when carrying a plasmid containing bexR ( data not shown ) , suggesting that native regulation of bexR is necessary for bistable PA1202 expression . Quantification of the frequency at which this switch in expression state occurs reveals a relatively infrequent switch with a bias in favor of the OFF to ON transition ( Table 1 ) . To determine whether bexR , like PA1202 , is expressed in a bistable manner , we constructed a reporter strain in which the putative bexR promoter region was placed upstream of a GFP-lacZ reporter in single copy at the ΦCTX attachment site in the PAO1 chromosome ( Figure 1D ) [16] , [17] . Individual cells of wild-type PAO1 carrying this PbexR-GFP-lacZ reporter either express the GFP reporter , or do not , leading to heterogeneity in the cell population ( Figure 1E ) . Interestingly , cells lacking BexR exhibit constitutively low-level expression of the reporter , suggesting that bistable expression from the bexR promoter also depends on BexR . Bistable expression from the bexR promoter was also observed at the colony level , suggesting long-term maintenance of the BexR expression state ( Figure S1 ) . The frequency of switching between expression states is similar for bexR and PA1202 , further supporting the hypothesis that bistable expression of bexR is upstream of PA1202 bistability ( Table 1 and Table S1 ) . Truncation of the bexR upstream sequence indicated that a 195 bp fragment of upstream DNA is still sufficient to drive expression of a lacZ reporter ( integrated in single copy in the chromosome ) when bexR is expressed from a plasmid , whereas an 88 bp fragment is not ( Figure 1F ) . Thus , the 195 bp region of DNA immediately upstream of bexR presumably contains the bexR promoter and BexR binding site ( s ) . Thus , BexR positively regulates expression of PA1202 and of its own gene , and bexR is itself bistably expressed , suggesting that other BexR target genes may also be expressed in a bistable manner . To determine the full extent of the BexR regulon in PAO1 , we compared the mRNA content of PAO1 ΔbexR cells containing either a bexR expression vector or an empty vector in both mid-logarithmic and stationary phases of growth using DNA microarrays . A total of 71 genes exhibited between a 2- and 70-fold change in expression , with most genes upregulated by ectopic expression of bexR ( Figure S2 ) . PA1202 was upregulated 70-fold upon ectopic expression of bexR in mid-logarithmic phase . Several genes downstream of PA1202 were also strongly upregulated by ectopic expression of bexR , suggesting that these comprise a BexR-regulated operon . This putative operon includes PA1203 , which is predicted to encode a redox protein , PA1204 , which is predicted to encode a NADPH-dependent FMN reductase , and PA1205 , which is predicted to encode a homolog of pirin , a widely conserved protein with oxygenase activity [18] . PA2698 , which is also predicted to encode a hydrolase , was upregulated 7-fold by ectopic expression of bexR , suggesting that a cohort of several enzymes are coordinately regulated by BexR . Several multidrug efflux pumps appeared to be regulatory targets of BexR , as downregulation of mexEF-oprN by 6- to 10-fold and upregulation of mexGHI-opmD by 7- to 13-fold was observed during ectopic expression of bexR . Several quorum sensing-regulated genes encoding secreted proteins were also positively regulated by ectopic bexR expression , such as PA0572 , which encodes a LasR-regulated Xcp secretion substrate with a predicted Zn-metalloprotease motif [19]–[21] . Finally , the LasR-regulated genes aprX , aprE , aprF and aprA , which encode components of the alkaline protease production and secretion machinery , were positively regulated by BexR . aprA , which encodes the alkaline protease precursor protein , plays a role in virulence in a Drosophila melanogaster orogastric model of pseudomonal infection , where it is thought to protect P . entomophila from antimicrobial peptides [22] . These results suggest that BexR controls the expression of a diverse set of genes , including some that encode predicted enzymes and others that encode quorum-regulated secreted proteases . Because bexR is itself bistably expressed we would predict that the expression of BexR target genes in wild-type cells should co-vary with the bexR expression state . To test this prediction , we isolated mRNA from cultures of wild-type attB::PbexR-lacZ OFF , attB::PbexR-lacZ ON and ΔbexR attB::PbexR-lacZ reporter strains at both mid-logarithmic and stationary growth phases and profiled relative transcript abundance by qRT-PCR . We observed an approximately 10-fold difference in abundance of bexR transcripts between OFF and ON cultures in mid-logarithmic phase , and an approximately 6-fold difference between OFF and ON cultures in stationary phase ( Figure 2A ) . Consistent with the idea that BexR target genes are expressed in a bistable manner in wild-type cells , expression of members of the putative PA1202 operon , from PA1202 to PA1205 , all co-varied with bexR expression ( Figure 2B ) , as did PA0572 and aprA ( though for aprA the difference in transcript abundance between ON and OFF cultures was only 2-fold ) ( Figure 2C ) . We were unable to observe significant bistable expression of the other apr genes , possibly due to the relatively modest effect of BexR on their expression . The abundance of the lasA transcript was not significantly different between ΔbexR and wild-type cultures , suggesting that the observed bistability of aprA and PA0572 ( which , like lasA , are LasR-regulated [19] ) is not due to differences in LasR function between ON and OFF cultures ( Figure 2C ) . Microarray analysis of cells ectopically expressing bexR suggests that two operons encoding multidrug efflux pumps are reciprocally regulated by BexR ( Figure S2 ) . However , this was not observed in wild-type cells in the OFF and ON states ( data not shown ) . Taken together , our data indicate that BexR is responsible for coordinate bistable expression of a variety of genes in wild-type P . aeruginosa , including two that encode quorum sensing-regulated secreted proteases ( PA0572 and aprA ) . To address whether BexR directly regulates transcription of its target genes , we used chromatin immunoprecipitation ( ChIP ) . We constructed a strain in which the native chromosomal copy of the bexR gene has been modified to encode a version of BexR containing a vesicular stomatitis virus glycoprotein ( VSV-G ) epitope tag at its C-terminus ( BexR-V ) . This strain retained the ability to bistably express PA1202 lacZ on LB agar containing X-Gal , suggesting that the VSV-G epitope tag does not interfere with BexR activity ( data not shown ) . We immunoprecipitated BexR-V-associated DNA from wild-type ON cultures grown to both mid-logarithmic and stationary phase and quantified occupancy of BexR-V at candidate target promoters relative to a control region not expected to bind BexR-V . BexR-V strongly occupies its own promoter , as well as those of PA1202 and PA0572 ( Figure 3 ) . Furthermore , BexR-V occupied the aprX and aprA promoters , but not the intervening DNA upstream of aprD . This suggests that BexR-V has at least two distinct binding sites within the apr locus . All occupancies were significantly higher than those observed in both wild-type OFF cultures and in a non-epitope tagged control strain ( Figure S3 ) . These results suggest that BexR regulates many of its target genes directly . The evidence presented thus far suggests that bexR encodes a bistably expressed transcription regulator that positively regulates its own expression . This is reminiscent of the competence switch in B . subtilis . In this system , ComK , the master regulator of competence , positively regulates transcription of its own gene , thereby enabling a non-linear response to increasing concentrations of ComK , which leads to bistability in the development of competence . Using single-cell fluorescent reporter analysis , it has been shown that the ComK positive feedback loop is required for bistable expression of competence [14] , [15] . We hypothesized that , in a similar manner , the positive feedback loop controlling bexR expression is required for bistable expression of the BexR regulon ( i . e . positive feedback of bexR creates a condition of hypersensitivity to variation in levels of BexR protein ) . If this hypothesis is correct a gradual increase in basal bexR expression should increase the proportion of ON relative to OFF cells specifically in a strain with an intact positive feedback loop . In a strain that lacks this positive feedback loop , a graded increase in bexR expression should lead to a corresponding increase in expression of bexR-regulated genes with no detectable bistability . Wild-type P . aeruginosa cells containing a PbexR-GFP-lacZ reporter construct integrated in single copy into the chromosome can be seen to exhibit BexR-dependent bistable expression of this reporter by fluorescence microscopy ( Figure 1E ) . Consistent with this observation , quantification of the fluorescence of individual cells within a culture derived from either an ON colony or an OFF colony reveals that cells in the ON and OFF expression states can be distinguished from one another , and that each culture contains both ON and OFF cells ( Figure 4 ) . To analyze the effect of positive feedback on bexR bistability , we constructed a pair of strains containing the PbexR-GFP-lacZ reporter construct and an isopropyl-β-D-thiogalactoside ( IPTG ) -inducible copy of bexR ( also provided in single copy from the chromosome from a different locus ) . One of these strains contained an unmarked , in-frame deletion of bexR ( the minus feedback strain , Figure 5A ) , whereas the other contained wild-type bexR at its native locus ( the plus feedback strain , Figure 5B ) . In the absence of IPTG , only cells of the reporter strain with the intact positive feedback loop displayed bistability , and contained two populations of cells corresponding to those in the ON and OFF expression states ( manifest in Figure 5B [and Figure 6B] as a population of cells with an essentially bimodal distribution of fluorescence intensities ) . Furthermore , a gradual increase in ectopically expressed bexR resulted in an increase in the proportion of ON relative to OFF cells only in the plus feedback strain ( Figure 5B ) ; in the strain lacking the positive feedback loop , cells responded relatively uniformly to increasing synthesis of ectopically expressed bexR ( manifest in Figure 5A as a population of cells with a normal distribution of fluorescence intensities , whose average fluorescence intensity increases with IPTG concentration ) . Importantly , for IPTG concentrations at which the average cell fluorescence intensity was similar between cells with and without feedback , two distinct cell populations ( ON and OFF ) were observed only in cells with an intact positive feedback loop ( Figure 5 ) . In particular , cells of the plus feedback strain at 0 . 5 mM IPTG had a mean fluorescence intensity of 1814 arbitrary units , which is similar to the mean fluorescence intensity of 1720 arbitrary units exhibited by the minus feedback strain at 4 mM IPTG . Whereas the mean reporter gene expression of these two cell populations , and thus the average abundance of BexR protein per cell , was quite similar under these two conditions , the existence of two subpopulations of cells occurred only in the presence of bexR autoregulation ( Figure 5 ) . These results suggest that positive feedback of bexR is necessary for bistability . Feedback-mediated bistable systems often exhibit a capacity for history-dependent behavior , or hysteresis [reviewed in 23] . Systems exhibiting hysteretic behavior may have different responses under identical conditions , depending on the conditions previously experienced . For example , in bistable expression of the lac operon of E . coli at low concentrations of a non-metabolizable lactose analog , the concentration of inducer at which initially uninduced cells turn on is higher than that at which initially induced cells turn off [24] , [25] . The behavior of this system at concentrations of inducer between these thresholds therefore depends on conditions previously encountered . Thus , systems with positive feedback can exhibit memory of previous expression states . To investigate the possibility that positive feedback of bexR can impart a memory of previous expression states on the system , we utilized the plus and minus feedback strains described above ( Figure 5 ) and observed their response over time to a pulse of ectopically expressed bexR , induced by a 2 hour exposure to 20 mM IPTG . In cells without an intact positive feedback loop , the IPTG pulse was sufficient to raise the mean fluorescence intensity to the level seen in wild-type ON cells ( Figure 6A and Figure 4 ) . However , this degree of expression from the PbexR-GFP-lacZ reporter was quickly lost upon removal of IPTG and subculturing of cells into fresh media . In contrast , cells of the plus feedback strain maintained their induced state for many generations after the removal of IPTG , suggesting that a brief period in which cells experience a high intracellular concentration of BexR is sufficient to induce a long-lasting ON state ( Figure 6B ) . Indeed , a pulse with IPTG for only 30 minutes is sufficient to induce a transition to a sustained ON state in the plus feedback strain ( Figure S4 ) . Only after 31 generations following removal of IPTG , do a portion of the cells begin to transition to the OFF state ( Figure 6B ) . Taken together , the results of our single-cell population analyses suggest a mechanism in which variation in basal expression of bexR in OFF cells is amplified by a positive feedback loop in a stochastically determined subset of cells that then transitions to the ON state and is maintained in that state by continued autoactivation of BexR ( Figure 7 ) .
Bistability is a mechanism by which bacteria can introduce phenotypic heterogeneity within an isogenic population , thereby creating a subset of cells capable of surviving the onset of an otherwise lethal situation . For example , some bacteria have the ability to survive antibiotic treatment without evolving bona fide resistance by stochastically entering a dormant “persister” state during vegetative growth [26] . A recent study suggests that a bexR transposon mutant has 2-fold increased sensitivity to the fluoroquinolone antibiotic ciprofloxacin , which is used in treatment of P . aeruginosa infections in CF patients , though the potential mechanism for this increased sensitivity was not addressed [27] , [28] . Although our findings raised the possibility that bistable bexR expression might lead to heterogeneity in ciprofloxacin resistance , we found no evidence that bexR contributed to the resistance of P . aeruginosa to ciprofloxacin , at least in strain PAO1 ( data not shown ) . Bistable expression of virulence factors has been previously reported in P . aeruginosa . For instance , the Type III secretion system is only expressed in a subset of cells grown in inducing conditions [29] . Additionally , the cupA fimbrial gene cluster is bistably expressed by P . aeruginosa when grown in anaerobic conditions [5] . Bistable expression of several virulence factors independently of one another may create several subtypes of cells with differing virulence potential within an isogenic population of infecting bacteria . Thus , bistable expression of virulence factors may represent a strategy employed by P . aeruginosa to generate cell types specialized to survive within different niches in the host . In P . entomophila , AprA has a significant role in virulence in a D . melanogaster oral model of infection , where it is thought to protect the bacterium from the effects of host-produced antimicrobial peptides [22] . Although oral models of D . melanogaster infection with P . aeruginosa have been used to successfully characterize bacterial virulence , these models have not been used to test the role of AprA in P . aeruginosa virulence [30] , [31] . If alkaline protease does play a role in defense against antimicrobial peptides in P . aeruginosa , upregulating aprA ∼2-fold in a subset of cells through BexR-mediated bistability may preemptively adapt a portion of the cell population to the sudden introduction to a particular host environment . P . aeruginosa alkaline protease has been shown to degrade a variety of human proteins and tissues and inhibit immune cell function , presumably by acting at the cell surface to modify phagocytic and chemotactic receptors ( reviewed in [32] ) . Alkaline protease has also been suggested to play a role in corneal keratitis [33] , although this role for AprA has been disputed more recently by the comparison of isogenic mutant strains [34] . However , our observation that wild-type strains of P . aeruginosa bistably express aprA may complicate the interpretation of earlier work . Interestingly , the rhizobacterium P . brassicacearum exhibits phenotypic variation in expression of an alkaline protease homolog , though whether this is mediated by bistability of a BexR homolog is unknown [35] . It has been suggested that heterogeneous production of extracellular proteases by an isogenic population of bacteria is an example of cooperative or altruistic behavior , as these proteases diffuse freely through the growth medium and can equally benefit all members of the population [36] . Thus , bistable production of alkaline protease or PA0572 , a predicted protease , may serve to benefit both ON and OFF cells in a population . Whether bistable expression of aprA , or other members of the BexR regulon , has a role in mammalian virulence remains to be seen . In contrast with aprA , many other regulatory targets of BexR are poorly characterized hypothetical genes . BexR-mediated bistability does not appear to be limited to P . aeruginosa PAO1 , as the homolog of PA1202 in P . aeruginosa PA14 , a more virulent clinical strain , is also bistably expressed in a BexR-dependent manner ( Figure S5 ) . This conservation across diverse strains of P . aeruginosa suggests an important biological role for BexR-mediated bistability . In this regard , a particularly interesting target of BexR is the PA1202 operon , which is strongly positively regulated by BexR . Several genes in this operon , such as PA1202 and PA1205 , are predicted to encode enzymes with catabolic activity directed against small molecules . This may point to a role for the BexR regulon in the ability of P . aeruginosa to metabolize and thereby detoxify certain small molecules . Co-regulation of a diverse set of genes by BexR may indicate that it is involved in manifestation of more than one phenotype . That these genes are expressed in a bistable manner suggests that their expression or repression may be detrimental to growth under certain conditions . We propose that positive feedback of bexR provides a mechanism for amplification and propagation of cell-to-cell variability in BexR levels . This regulatory circuit is similar to the one governing competence development in B . subtilis . Experiments in this system have suggested that noisy expression of comK results in ComK levels in a subpopulation of cells crossing a threshold level for comK autoactivation , causing differentiation into the competent state [14] , [15] , [37] . Noise in bexR expression may also provide the basis for generating cell-to-cell variability in BexR levels . The frequency of the BexR switch differs from that of the ComK switch . Whereas B . subtilis has been directly observed to enter a competent state in approximately 3 . 6% of cell division events [38] , P . aeruginosa enters into the BexR-ON state approximately 10-fold less frequently , and the BexR-OFF state even less so ( Table 1 and Table S1 ) . These low frequencies are on par with classical phase-variation systems , but in the case of BexR , the expression state stability appears to be epigenetically mediated . This low switching frequency may be a function of the high degree of hysteresis observed in the BexR switch . Biological systems capable of hysteretic behavior can retain a memory of previous exposure to inducing conditions , and this has been observed in both naturally occurring and synthetic systems [25] , [39] . Strictly speaking , hysteresis is not a necessary characteristic of bistable systems , as a synthetic feedback-mediated bistable system was observed to exhibit clear bistability but display no history-dependent response [40] . Nevertheless , hysteresis is often associated with bistable systems , and that it is observed in the BexR switch may suggest that retaining memory of previous conditions is beneficial to the cell . In B . subtilis , regulation of ComK levels is achieved by degradation of ComK by the MecA/ClpCP complex and the inhibition thereof by ComS [41] , [42] . Our single-cell population analyses indicate that directly modulating BexR levels by induction of ectopic synthesis can affect the frequency at which cells differentiate into the ON state ( Figure 5 ) . Modulation of BexR levels or activity in wild-type cells may provide a mechanism for fine-tuning the dynamics of this bistability . There may be accessory factors , perhaps themselves BexR-regulated , that affect BexR levels or activity . A mechanism for modulating BexR autoactivation dynamics may allow P . aeruginosa to regulate switching frequency in response to external conditions . As LysR-type transcription activators often bind to small molecules to alter their DNA-binding and regulatory properties , it is possible that the dynamics of the BexR switch may be tunable by a coinducer molecule [43] . However , no such molecule has yet been identified . The results presented here outline a model for differentiation into the BexR-ON state , but do not address the mechanism by which a BexR-ON cell can revert to the BexR-OFF state . Previous studies suggest that escape from a positive feedback loop is often mediated by an accessory process . For example , escape from competence in B . subtilis occurs when reduction in ComS levels promotes ComK proteolysis by the MecA/ClpCP complex , relieving ComK autoactivation [38] . The switch from BexR-ON to BexR-OFF may also involve some antagonistic process . Unlike several other feedback-mediated bistable switches , the switch from ON to OFF in the case of BexR appears to occur only in a stochastically determined subset of cells . For example , escape from competence in B . subtilis occurs because comS transcription is repressed by ComK in the competent state and ComS protein gradually depletes in all cells [38] . In contrast , the BexR-ON state is relatively stable and heritable , and is lost only in a subpopulation of cells . The existence of a stochastic process mediating the switch to BexR-OFF that is distinct from the one mediating the switch to BexR-ON , is further supported by the ∼60-fold directional bias in switching frequencies ( Table 1 ) . This process may take the form of transcription regulation of bexR or post-translational modulation of BexR levels or activity , and we are currently investigating these possibilities .
P . aeruginosa strains PAO1 and PA14 were provided by Arne Rietsch ( Case Western Reserve University ) . E . coli DH5α F'IQ ( Invitrogen ) was used as the recipient strain for all plasmid constructions , whereas E . coli strain SM10 ( λpir ) was used to mate plasmids into P . aeruginosa . The PA1202 lacZ reporter strain ( PAO1 PA1202 lacZ ) contains the lacZ gene integrated immediately downstream of the PA1202 gene on the PAO1 chromosome and was made by allelic exchange . PCR products 486 bp and 513 bp in length flanking the 3′ end of PA1202 were amplified and spliced together to add KpnI , NcoI and SphI sites two bases after the PA1202 stop codon . This PCR product was cloned as a SacI/PacI fragment into pEXG2 [44] . The lacZ gene was subsequently cloned into this construct as a KpnI/SphI fragment derived from pP18-lacZ ( Arne Rietsch , unpublished work ) , generating plasmid pEXF1202-lacZ . This plasmid was then used to create reporter strains PAO1 PA1202 lacZ and PA14 PA1202 lacZ by allelic exchange . The deletion construct for the bexR gene ( PA2432 ) was generated by amplifying regions 398 bp and 360 bp in length that flank bexR in the PAO1 genome by the PCR and then splicing the flanking regions together by overlap extension PCR; deletions were in-frame and contained the 9-bp linker sequence 5′-GCGGCCGCC-3′ . The resulting PCR product was cloned on an EcoRI/BamHI fragment into plasmid pEX18Gm [45] , yielding plasmid pEXM2432 . This plasmid was then used to create strains PAO1 ΔbexR , PAO1 PA1202 lacZ ΔbexR and PA14 PA1202 lacZ ΔbexR by allelic exchange [45] . Deletions were confirmed by the PCR . The attB::PbexR-lacZ reporter strains contain fragments of the bexR promoter fused to the lacZ gene and integrated in single copy into the attB locus in the PAO1 chromosome and were made by site-specific integration followed by backbone excision through transient synthesis of FLP recombinase from plasmid pFLP2 [17] , [45] . PCR products spanning from 91 , 198 or 297 bp to 3 bp upstream of the bexR start codon were amplified and cloned as EcoRI/XhoI fragments into mini-CTX-lacZ [17] , which contains a consensus Shine-Dalgarno sequence upstream of lacZ , yielding plasmids mini-CTX-PF2432-lacZ . 1 , mini-CTX-PF2432-lacZ . 2 and mini-CTX-PF2432-lacZ . 3 , respectively . These plasmids were then used to create reporter strains PAO1 attB::PbexR . 1-lacZ , PAO1 ΔbexR attB::PbexR . 1-lacZ , PAO1 attB::PbexR . 2-lacZ , PAO1 ΔbexR attB::PbexR . 2-lacZ , PAO1 attB::PbexR . 3-lacZ and PAO1 ΔbexR attB::PbexR . 3-lacZ . An EcoRI/XhoI fragment of mini-CTX-PF2432-lacZ . 3 was subcloned into mini-CTX-GFP-lacZ [16] , yielding plasmid mini-CTX-PF2432-GFP-lacZ . 3 . This plasmid was then used to create the fluorescent reporter strains PAO1 attB::PbexR . 3-GFP-lacZ and PAO1 ΔbexR attB::PbexR . 3-GFP-lacZ . The BexR-VSV-G integration vector was generated by first cloning a PCR-amplified DNA fragment containing ∼300 bp of sequence from the 3′ portion of the bexR gene on a HindIII/NotI fragment into plasmid pP30Δ-YTAP [4] , generating plasmid pP30Δ-BexR-TAP . This HindIII/NotI fragment was then subcloned into pP30ΔFRT-MvaT-V [46] , generating plasmid pP30ΔFRT-BexR-V . This plasmid was used to make strain PAO1 PA1202 lacZ BexR-V by homologous recombination at the bexR locus followed by backbone excision through transient synthesis of FLP recombinase from plasmid pFLP2 [45] . Production of the BexR-V protein was confirmed by Western blotting with an anti-VSV-G antibody ( Sigma ) . Plasmid pBexR is a derivative of pPSV35 [44] and directs the synthesis of the BexR protein under control of the IPTG-inducible lacUV5 promoter . The plasmid was made by subcloning an EcoRI/HindIII DNA fragment containing a consensus Shine-Dalgarno sequence and the bexR gene into pPSV35 . The attTn7::TOPLAC-bexR strains contain a construct which directs the synthesis of the BexR protein under control of the IPTG-inducible TOPLAC promoter stably integrated into the genome in single copy at the attTn7 locus . The TOPLAC promoter in this construct is a derivative of the lac promoter that contains two lac operator sequences centered at positions −63 . 5 and +11 . The sequence of this promoter is 5′-CACTACGTGCTCGAGGGTAAATGTGAGCACTCACAATTTATTCTGAAATGAGCTCTTTACACGTCCTGCTGCCGGCTCGTATGTTGTGTGGAATTGTGAGCGGATAACAATTAAGCTTAGTCGACAGCTAGCCGGATCC-3′ , where the -35 and -10 sequences are underlined and the lac operator sequences are shown in bold . The bexR gene is inserted downstream of the TOPLAC promoter with a consensus Shine-Dalgarno sequence . This construct was inserted between the ends of the Tn7 transposon on pUC18-mini-Tn7T-LAC [47] , generating plasmid pUC18-mini-Tn7T-TOPLAC-bexR . This plasmid was used to make strains PAO1 attB::PbexR . 3-GFP-lacZ attTn7::TOPLAC-bexR and PAO1 ΔbexR attB::PbexR . 3-GFP-lacZ attTn7::TOPLAC-bexR by site-specific recombination [47] . Cells were grown with aeration at 37°C to mid-logarithmic phase in LB supplemented as needed with gentamicin ( 25 µg/ml ) and IPTG ( 0 . 1 mM ) . Cells were permeabilized with sodium dodecyl sulfate and CHCl3 and assayed for β-galactosidase activity as described previously [48] . Assays were performed at least twice in triplicate on separate occasions . Representative data sets are shown . Cultures of PAO1 ΔbexR attB::PbexR . 3-lacZ and PAO1 attB::PbexR . 3-lacZ in the OFF and ON states were inoculated in quadruplicate at starting OD600 of ≈0 . 01 and grown with aeration to an OD600 of ≈0 . 55 ( representing mid-logarithmic phase ) and to an OD600 of ≈2 . 4 ( representing stationary phase ) at 37°C in LB . Cells were then harvested by centrifugation and RNA prepared essentially as described [49] . Transcripts were quantified by quantitative real-time RT-PCR as described [50] . Switching-frequency calculations were performed essentially as described [51] , except that cells were plated on LB agar plates containing 50 µg/ml X-Gal and grown at 37°C . Error values represent 1 standard deviation ( SD ) from the mean switching frequency . Cultures of PAO1 ΔbexR containing plasmid pPSV35 [44] or pBexR were grown with aeration at 37°C in LB containing gentamicin ( 25 µg/ml ) . Triplicate cultures of each strain were inoculated at a starting OD600 of ≈0 . 01 and grown to an OD600 of ≈0 . 5 ( representing mid-logarithmic phase ) and to an OD600 of ≈2 . 3 ( representing stationary phase ) . RNA isolation , cDNA synthesis , and cDNA fragmentation and labeling were performed essentially as described previously [49] . Labeled samples were hybridized to Affymetrix GeneChip P . aeruginosa genome arrays ( Affymetrix ) . Data were analyzed for statistically significant ( p<0 . 05 , fold change >2 ) changes in gene expression using GeneSpring GX . Cultures of PAO1 PA1202 lacZ BexR-V in either the ON or OFF state were inoculated in quadruplicate at a starting OD600 of ≈0 . 01 and grown with aeration to an OD600 of ≈0 . 5 ( representing mid-logarithmic phase ) and to an OD600 of ≈2 . 0 ( representing stationary phase ) at 37°C in LB . ChIP was then performed with 3 ml of culture and fold enrichment values were measured by quantitative PCR relative to the PA2155 promoter essentially as described [46] . For fluorescence micrograph analysis , cultures were fixed with formaldehyde and glutaraldehyde at 2 . 4% and 0 . 04% , respectively , and cells were allowed to fix for 30 minutes at room temperature . Cells were washed three times with PBS and imaged on a Nikon TE2000 inverted microscope outfitted with a Nikon Intensilight illuminator , a Coolsnap HQ2 charge-coupled device camera from Photometrics and a Nikon CFI Plan Apo VC ×100 objective lens ( 1 . 4 NA ) for differential interference contrast ( DIC ) imaging . For GFP images the ET-GFP filter set ( Chroma 49002 ) was used . Images were captured using Nikon Elements software , which was also used for quantification of fluorescence in individual cells . This was done by automatically defining cell boundaries using the DIC image , excluding cells that were poorly focused , narrower than 0 . 5 µm , longer than 4 . 0 µm or shorter than 0 . 5 µm , and using those regions to quantify the GFP image . Values given are subtracted for background fluorescence . At least 400 cells were imaged for each timepoint , and the fluorescence intensities of a random subset of 250 cells are displayed in scatter plots . Images were exported to Adobe Photoshop CS4 for preparation . For the hypersensitivity experiment ( Figure 5 ) , cells were grown with aeration at 37°C to mid-logarithmic phase in LB supplemented as needed with IPTG and prepared for microscopy as described above . The experiment was performed at least twice in duplicate on separate occasions . A representative data set from a single replicate is shown . The hysteresis experiment ( Figure 6 ) was performed by growing cells with aeration at 37°C in LB and either treating them with 20 mM IPTG for 2 hours or 30 minutes immediately before reaching mid-logarithmic phase , or not treating them with IPTG . A sample was then taken and prepared for microscopy ( corresponding to the 0 generation time point ) as described above while the remaining cells were washed with LB to remove the IPTG , and inoculated into fresh media at a 1∶4 dilution . Cells were then grown continuously for 2 generations to mid-logarithmic phase , a sample was taken and prepared for microscopy ( corresponding to the 2 generation time point ) and a fresh culture was inoculated at a 1∶16 dilution with the remaining cells . Cells were then grown continuously for 4 generations to mid-logarithmic phase , a sample was taken and prepared for microscopy ( corresponding to the 6 generation time point ) and a fresh culture was inoculated at a 1∶16 dilution with the remaining cells . Cells were then grown continuously for 5 generations to late-logarithmic phase , a sample was taken and prepared for microscopy ( corresponding to the 11 generation time point ) and a fresh culture was inoculated at a 1∶32 dilution with the remaining cells . Cells were then grown continuously for 5 generations to late-logarithmic phase and a fresh culture was inoculated at a 1∶16 dilution . Cells were then grown continuously for 3 generations to mid-logarithmic phase , a sample was taken and prepared for microscopy ( corresponding to the 19 generation time point ) , remaining cells were allowed to grow for 1 . 5 generations to early stationary phase and used to inoculate a fresh culture at a 1∶100 dilution . Cells were then grown continuously for 7 generations ( overnight ) and used to inoculate a fresh culture at a 1∶100 dilution . Cells were then grown continuously for 3 . 5 generations to mid-logarithmic phase and a sample was taken and prepared for microscopy ( corresponding to the 31 generation time point ) . The experiment was performed at least three times in duplicate on separate occasions . A representative data set from a single replicate is shown . | Bistable switches allow the expression of a gene , or set of genes , to switch from one stable expression state to another and can generate cells with different phenotypes in an isogenic population . In this work we uncover a previously unidentified bistable switch that controls virulence gene expression in the opportunistic pathogen P . aeruginosa . This switch is controlled by a LysR-type transcription regulator that we call BexR . As well as identifying specific genes that are regulated by BexR , we show that bexR is itself bistably expressed and positively autoregulated . Furthermore , we present evidence that positive autoregulation of bexR is necessary for bistable expression of the BexR regulon . Our findings support a model for BexR-mediated bistability in which positive feedback amplifies bexR expression in a stochastically determined subset of cells , giving rise to heterogeneous expression of BexR target genes within a cell population . By generating diversity in an isogenic population of P . aeruginosa this bistable switch may ensure the survival of a subset of cells in adverse conditions , such as those encountered in the host . Our study defines an epigenetic mechanism for phenotypic variation in P . aeruginosa . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"infectious",
"diseases/nosocomial",
"and",
"healthcare-associated",
"infections",
"genetics",
"and",
"genomics/gene",
"expression",
"infectious",
"diseases/bacterial",
"infections",
"genetics",
"and",
"genomics/epigenetics",
"microbiology/microbial",
"growth",
"and",
"development"
] | 2009 | Epigenetic Control of Virulence Gene Expression in Pseudomonas aeruginosa by a LysR-Type Transcription Regulator |
A promising alternative to address the problem of acquired drug resistance is to rely on combination therapies . Identification of the right combinations is often accomplished through trial and error , a labor and resource intensive process whose scale quickly escalates as more drugs can be combined . To address this problem , we present a broad computational approach for predicting synergistic combinations using easily obtainable single drug efficacy , no detailed mechanistic understanding of drug function , and limited drug combination testing . When applied to mutant BRAF melanoma , we found that our approach exhibited significant predictive power . Additionally , we validated previously untested synergy predictions involving anticancer molecules . As additional large combinatorial screens become available , this methodology could prove to be impactful for identification of drug synergy in context of other types of cancers .
Targeted therapies designed to specifically target molecules involved in carcinogenesis have achieved remarkable antitumor efficacy . In melanoma , over half of patients are reported to harbor activating mutations in the BRAF oncogene [1 , 2] . BRAF inhibitors , such as vemurafenib , have been developed to selectively kill mutant BRAF positive cells [3] . Patients initially exhibited significant responses to these drugs , with 48% responding to vemurafenib in phase 1 and 2 clinical trials , however resistance developed within months [3] . Combination therapy has been proposed for preventing and overcoming resistance . This is thought to be a promising option because resistance to the combinatorial therapy would require either acquisition of multiple mutations rapidly [4] or an individual mutation that is able to bypass both drugs [5] , both of which are low probability events . Additional goals of combination therapy are to lower drug dosage levels in order to reduce the frequency and severity of adverse events and to achieve enhanced effectiveness through either drug additivity or synergy [4] . Drug synergy can occur through a variety of mechanisms . These include enhancement of bioavailability , through inhibition of parallel pathways[6] , and chemosensitization , in which the first compound primes the cells to be sensitive to the second drug[7] . Synergy is generally quantified through either effect based or dose-effect based methodologies . Effect-based methods compare the independent effects of drugs , while dose-effect based methods assume nonlinear individual dose–effect curves[8] . The most popular effect based method is the Bliss Independence model , which assumes that drugs act independently and the expected additive effect is based on the common probabilistic independence formula[8] . However limitations of this approach include that it does not account for nonlinearity in dose response curves [9] and its independence assumption . Since the mechanism of action for many drugs remain unknown , the validity of the independence assumption is often not met [8] . A popular dose-effect based method is the Chou-Talalay Combination Index ( CI ) , which is a median-effect equation based on the “mass-action law”[10] . A major limitation of the Chou-Talalay method is its dependence on accurate and well-defined dose-effect curves , which are not always available[8] . Given the poor prognosis of BRAF melanoma and the rapid rate at which resistance develops and tumors progress , there is an urgent need to identify suitable combinations . The first drug combination for treatment of advanced melanoma was approved in January 2014 and involved the BRAF inhibitor , dabrafenib , and the MEK inhibitor trametinib . This combination was pursued due to the great response rates of the individual drugs with the goal of preventing drug resistance . Indeed , it has been observed that the combination delays the development of resistance and prolongs progression free and overall survival[11] . However it is certain that many more combinations exist and are not yet known . Additionally , a subset of patients that were treated with the combination of dabrafenib and trametinib have developed resistance to this combination therapy[11] . Existing methods that have been developed to predict synergistic combinations have generally relied on mechanistic insights . However they have been applied only in limited specific contexts , such as in B cells [7] . Furthermore numerous studies have shown that synergy is very dependent on context [2 , 6 , 7] . This makes it difficult to utilize any prior knowledge about synergistic drug combinations from other different cancers or genotypes . A systematic method for identifying optimal combinations would therefore be highly impactful . Here we propose a computational approach utilizing existing high-throughput drug screen data to help identify other combinations that are both synergistic and effective in the context of mutant BRAF melanomas .
We set out to determine whether combination efficacy and synergy could be predicted from single agent efficacies . Previous computational approaches to drug combinations have shown that the dose response curves of single agents exhibit predictive power for identifying synergistic combinations [7] . To further investigate this , we utilized a high-throughput drug screen that was performed by Held et al . [2] . In this study , the response of 150 single agents and a large combinatorial drug screen involving 40 drugs were experimentally tested across mutant BRAF , mutant RAS , and wild-type BRAF and RAS ( WT ) cell lines . For each drug pair , we derived a feature set that consisted of the mean and difference of the single agent dose response in each tested cell line . The single agent dose response was represented as the percent of concentration required to inhibit 50% of growth inhibition ( GI50 ) . Altogether , this resulted in a total of 54 total features representing the similarity of a drug pair’s efficacies in 27 melanoma cell lines ( S1 Fig ) . The results of the combinatorial drug screen were used to identify genotype-selective and synergistic combinations . Genotype-selective combinations were defined to be those that yielded an average 15% or greater growth inhibition exclusively in the genotypic group and achieved at least 50% growth inhibition within the genotypic group . We further defined a general effective combination to be one that achieved at least 70% growth inhibition . Finally we computed synergy labels for each combination using the Chou-Talalay synergy combination index ( CI ) metric . We decided to use the Chou-Talalay approach because the Chou-Talalay CI was provided in the original dataset and furthermore the mechanism for many drugs in the training set is unknown . We defined a synergistic combination to be a pair of drugs that demonstrated CI < -1 at any concentration level . Overall , the combinatorial screen helped identify 248 BRAF-selective and 161 synergistic combinations . We then trained random forest models [12] on 780 drug combinations for each of the outcomes described above in context of BRAF and RAS melanomas ( Fig 1A , S2 Fig ) . We evaluated our approach using 10-fold cross-validation and found that our model exhibits significant power ( Table 1 ) for predicting both synergy ( AUC = 0 . 8663 , Accuracy = 0 . 8213 ) and genotype-selective efficacy ( AUC = 0 . 8809 , Accuracy = 0 . 8230 ) in context of BRAF melanomas ( Fig 1B ) . Importantly both models maintained high specificity rates ( 0 . 9494 and 0 . 8894 for synergy and effectiveness respectively ) which suggests that there would be few false leads . As a control to identify limitations of our approach , we evaluated our model by making predictions for “sham combinations” , which were cases in which a drug is combined with itself . We found that 98% ( 147/150 ) of the sham combinations were predicted to not be synergistic . We also found that both the effectiveness and synergy models exhibit significant robustness ( Fig 1C ) . At 25% of the original number of combinations used in the training set , the BRAF-specific effectiveness approach maintained 77 . 56% accuracy , 89 . 27% specificity , and 54 . 91% sensitivity . This suggests that fewer combination testing could be performed while maintaining strong confidence in the positive predictions , given that the specificity remains high . Since high-throughput screens require significant resources and time , this type of approach could prove to be valuable in screening the larger space of drug combinatorial pairs given that a suitable representative set is chosen for initial testing . The single agent screens performed by Held et al [2] included 110 drugs that were not tested in the combinatorial screen , so we applied our approach to the 10 , 395 additional untested combinations . We predicted 842 combinations to be synergistic , 890 to be effective , and 304 to be both effective and synergistic in context of mutant BRAF melanoma . We found that our predictions had noticeable patterns of synergy and effectiveness ( S3A Fig ) . Predicted synergistic combinations involved drugs that had varying levels of efficacy across the different mutant BRAF cell lines , with synergistic combinations demonstrating a trend towards lower correlation of GI50 values across the mutant BRAF cell lines ( p = 0 . 07929 , Kolmogorov-Smirnov Test ) . In contrast , combinations involving drugs with similar efficacy profiles across the different cell lines were generally predicted to be non-synergistic . Thus it appears that our approach drew its strength from the large number of tested cell lines . To further evaluate our approach , we compared these predictions to an independent high-throughput screen that tested 5 , 778 combinations involving 108 drugs at two concentration levels , high and low[13] . Our prediction dataset contained 274 combinations that overlapped with this independent dataset . We found that our predicted effective combinations had a significantly higher growth inhibition levels than our predicted non-effective combinations ( p = 0 . 002602 , Student’s t Test ) ( S3B Fig ) . We next identified a subset of 7 drugs with a diverse set of predictions ( S3C Fig ) . These drugs included a BRAF inhibitor ( PLX4720 ) , a statin ( Simvastatin ) , two chemotherapies ( Doxorubicin , Paclitaxel ) , and three drugs of other various mechanisms ( Fak Inhibitor 14 , Gefitinib , 17AAG ) . We tested each of these drugs both alone and in combination in the mutant BRAF melanoma cell line MALME-3M at low , medium , and high concentrations , estimated from their GI10 , GI25 , and GI50 values respectively . We found that our method continued to demonstrate significant predictive power when tested on cell lines that were independent of the original training set ( Fig 2A , Table 2 ) . We validated 82% and 64% of the effectiveness ( Fig 2B ) and synergy predictions ( Fig 2C ) respectively . Importantly , we also found that the false discovery rates ( FDR ) for both synergy and effectiveness predictions remained relatively low ( 14 . 3% for synergy , 12 . 5% for effectiveness ) despite being tested in a different setting . BRAF inhibitors are of high interest for treating BRAF-mutant melanomas due to their selectivity and effectiveness . To further investigate the efficacy and synergy of combinations involving PLX4720 , we performed more extensive experiments for predicted synergistic and non-synergistic drug partners . PLX4720 was predicted to be synergistic with FAK inhibitor 14 and non-synergistic with 17AAG , a Hsp90 inhibitor . Instead we found both combinations to be synergistic when tested across the 9 combinations of varying concentrations . Interestingly , we observed that while 17AAG appeared to be synergistic when combined with PLX4720 held at a constant rate , PLX4720 itself was not synergistic when 17AAG was held constant ( Fig 3A ) . However PLX4720 was very synergistic when combined with a constant dosage of FAK Inhibitor 14 ( Fig 3B ) , consistent with the synergy prediction . Additionally we observed that PLX4720 was highly synergistic with paclitaxel . This combination represents a potentially impactful combination since paclitaxel and vemurafenib are both used in clinical trials for the treatment of melanoma [3 , 14] . Furthermore previous reports have suggested that the combination of BRAF inhibitors with paclitaxel represents a promising therapeutic approach for overcoming resistance in BRAF melanomas [15] .
We found that drug synergy and combinatorial effectiveness can be predicted from a relatively small subset of combinations based only upon single drug efficacies . We experimentally validated novel predictions involving 7 drugs in a BRAF mutant cell line with FDR<0 . 15 . This analysis included compounds that span a variety of drug classes , including targeted therapies and chemotherapies . Additionally , an analysis of the model robustness suggested that it is possible to confidently make these predictions with even smaller subsets , while maintaining significant confidence in the positive predictions . We note that while we propose that a smaller training set can be used to infer these combinations , it is critical to retain a distinctive and representative set of drug combinations in the training set . The classification errors , particularly those involving synergy predictions , may be due in part to varying genetic conditions since the combinations were tested in a different cell line than the original training set . This is supported by one combination that was tested both in our experiments and in the larger combinatorial drug screen that had inconsistent synergy and effectiveness levels . The combination of Fak Inhibitor 14 and gefitinib was found to be synergistic in the combinatorial screen , however we found it to be non-synergistic in our experiments . Consequently we believe that our false discovery rate would have been even lower if we had tested the combinations in the same setting that was used to generate the training set . However the validation of the majority of our predictions across slightly variable contexts is highly relevant for the treatment of patient cancers , in which the treated patient populations involve different individuals and thus have slightly different genomic profiles than the population under which the therapy was conceived . Additionally we would like to emphasize that any effectiveness ( or synergy ) predictions that our model makes are in the context of mutant BRAF cell lines . While we hope that some of these findings may be translatable to human patients , many other factors must first be considered . It is important to note that we did not consider maximum tolerated doses ( MTD ) in our analysis . A retrospective analysis revealed that there were two drugs included in our analysis that were tested at levels above MTD for humans: Obatoclax and Tamoxifen . Additionally there are many drugs included in the study do not have known MTD . This is because these drugs are experimental and thus have not yet had this evaluated in clinical trials[16] . While we do not believe that this biases the model , it does highlight the importance of clear model interpretation . In particular , we observed in our experiments that synergy generally did not occur at high dosage levels ( Fig 3 ) , which further suggests that the model would not be biased by including drugs that were tested above MTD . Importantly our approach allows subsequent experimental studies to be prioritized on promising combinations , which can be focused on more clinically relevant information such as MTD . The use of synergistic drug combinations has the potential to help prevent and overcome drug resistance . It is hypothesized that the application of drug combinations with initial treatment lower the odds of resistance occurring because it requires multiple mutations to bypass both drugs , which is a lower probability event than each drug individually[4 , 5] . This could be particularly impactful for combinations involving BRAF inhibitors , which individually have demonstrated remarkable responses in patients but suffer from the rapid development of resistance[3] . To further explore how models could be used to identify drug combinations that overcome drug resistance , we used our training set to derive a set of 24 effective and 12 non-effective combinations involving vemurafenib in resistant cell lines . We applied the framework of our approach to this small dataset and found that there is an underlying signal for predicting combinations that overcome resistance ( AUC = 0 . 677 , Accuracy = 0 . 697 ) . Thus approaches such as ours may be applicable both to directly predicting combinations that overcome resistance , as well as predicting combinations that may help prevent resistance from developing . Existing methods have previously found that the inclusion of features representing the biological mechanisms of the drug have been most successful[7] . However this information is often not available . Indeed only 50% ( 20/40 ) of the drugs in our training set have information about the drug’s target . Thus our method would likely improve as this type of information becomes more widely available enough to include in the model . However our model was able to exhibit significant predictive power despite this information not being available . We hypothesize that this was due to the large number of cell lines that each single agent was tested in . We found that generally the predicted synergistic combinations involved drugs that had varied single agent efficacies across the different mutant BRAF cell lines . While we have trained and tested our approach in the context of BRAF mutant melanoma , the approach itself is applicable to other types of cancers . As additional large combinatorial screens become available , this methodology could prove to be impactful for the identification of drug synergy within the larger universe of possible drug combinations .
150 Single agent and 780 combinatorial efficacies were obtained from the Held et al [2] study . The single agent efficacies were collapsed to their GI50 values , which is the concentration of the drug required to inhibit 50% of cell growth . Features representing a drug pair were constructed by taking the mean and difference ( S1 Fig ) of the GI50 values for each of 27 tested cell lines ( 15 mutant BRAF , 6 mutant RAS , 6 wtBRAF/wtRAS ) . The combinatorial results in 19 cell lines ( 8 mutant BRAF , 6 mutant RAS , and 5 WT ) were then used to construct labels for each of the 780 drug pairs . Genotype-selective combinations were defined to be those that yielded an average 15% or greater growth inhibition exclusively in the genotypic group and achieved at least 50% growth inhibition within the genotypic group and a general effective combination was defined to be one that achieved at least 70% growth inhibition . Synergy labels for each combination were determined using the Chou-Talalay synergy combination index ( CI ) metric . We defined a synergistic combination to be a pair of drugs that yielded a CI less than -1 at any concentration level . A random forest model was then trained on the above-described data and evaluated using 10-fold cross-validation . Predictions were made using the trained model for the 10 , 395 untested combinations that had single agent efficacy information in the dataset . | While targeted therapies have achieved remarkable antitumor responses , resistance to targeted agents frequently develops and renders the targeted drug ineffective . Combination therapies have been successful in delaying resistance and overcoming resistance . Additional goals of combination therapy are to achieve enhanced effectiveness through drug synergy and to reduce the frequency and severity of adverse events through lower individual drug dosage levels . However the identification of synergistic drug combinations is often a labor and resource intensive process . Therefore a systematic method for identifying optimal combinations would be highly impactful . Here we present a computational method for predicting synergistic and effective drug combinations using only single drug efficacy information . We have applied and validated our method on a high-throughput drug screen of 780 combinations involving 40 individual molecules in the context of mutant BRAF melanoma . Additionally we have made predictions and validated 11 previously untested drug combinations with a diverse set of outcomes . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"pharmacologic",
"analysis",
"medicine",
"and",
"health",
"sciences",
"antibiotic",
"susceptibility",
"testing",
"dose",
"prediction",
"methods",
"cancer",
"treatment",
"biological",
"cultures",
"cancers",
"and",
"neoplasms",
"drug",
"screening",
"simulation",
"and",
"modeling",
"oncology",
"systems",
"science",
"mathematics",
"pharmaceutics",
"cell",
"cultures",
"pharmacology",
"molecular",
"biology",
"techniques",
"melanoma",
"cells",
"research",
"and",
"analysis",
"methods",
"computer",
"and",
"information",
"sciences",
"agent-based",
"modeling",
"melanomas",
"high",
"throughput",
"screening",
"molecular",
"biology",
"molecular",
"biology",
"assays",
"and",
"analysis",
"techniques",
"synergy",
"testing",
"biology",
"and",
"life",
"sciences",
"cultured",
"tumor",
"cells",
"physical",
"sciences"
] | 2017 | A Computational Approach for Identifying Synergistic Drug Combinations |
Microsporidia are obligate intracellular parasites with the smallest known eukaryotic genomes . Although they are increasingly recognized as economically and medically important parasites , the molecular basis of microsporidian pathogenicity is almost completely unknown and no genetic manipulation system is currently available . The fish-infecting microsporidian Spraguea lophii shows one of the most striking host cell manipulations known for these parasites , converting host nervous tissue into swollen spore factories known as xenomas . In order to investigate the basis of these interactions between microsporidian and host , we sequenced and analyzed the S . lophii genome . Although , like other microsporidia , S . lophii has lost many of the protein families typical of model eukaryotes , we identified a number of gene family expansions including a family of leucine-rich repeat proteins that may represent pathogenicity factors . Building on our comparative genomic analyses , we exploited the large numbers of spores that can be obtained from xenomas to identify potential effector proteins experimentally . We used complex-mix proteomics to identify proteins released by the parasite upon germination , resulting in the first experimental isolation of putative secreted effector proteins in a microsporidian . Many of these proteins are not related to characterized pathogenicity factors or indeed any other sequences from outside the Microsporidia . However , two of the secreted proteins are members of a family of RICIN B-lectin-like proteins broadly conserved across the phylum . These proteins form syntenic clusters arising from tandem duplications in several microsporidian genomes and may represent a novel family of conserved effector proteins . These computational and experimental analyses establish S . lophii as an attractive model system for understanding the evolution of host-parasite interactions in microsporidia and suggest an important role for lineage-specific innovations and fast evolving proteins in the evolution of the parasitic microsporidian lifecycle .
Microsporidia are a diverse phylum of obligate intracellular parasites related to fungi . Over 1300 species have been described in approximately 160 genera and , as is the case for other microbial eukaryotes , a vast undescribed diversity is thought to exist in the environment [1] . Microsporidia are important pathogens of a broad range of animal groups: they can infect immunocompromized humans , such as those with HIV/AIDS , and are major pathogens of fish and invertebrates , representing a significant threat to sericulture [2] and fisheries [3] . Their unusual lifecycle has also attracted attention , particularly the unique mechanism by which microsporidia gain entrance to host cells . Outside the host cell , microsporidia exist as a resistant spore containing a coiled polar tube . Upon coming into contact with a host cell , or appropriate stimulus , the spore rapidly everts this tube , penetrating the host cell membrane and delivering the spore contents to the host cytoplasm , where proliferation and the next round of spore production occurs [4] . In addition to their importance as parasites of animals , Microsporidia have attracted much attention as eukaryotic model systems for reductive genome evolution . The 2 . 9 Mb genome of the microsporidian Encephalitozoon cuniculi was one of the first eukaryotic genomes to be sequenced [5] . Analysis of the E . cuniculi genome revealed a highly reduced and streamlined genome which had lost or simplified many biochemical pathways , had truncated genes , shortened intergenic spaces and had almost entirely lost introns and repetitive DNA [5]; its close relative Encephalitozoon intestinalis has an even smaller genome , at 2 . 3 Mb [6] . Interestingly , while microsporidian genomes are consistently smaller than those of their opisthokont relatives , there is a ten-fold difference in genome size within the phylum , with some genomes as large as 24 Mb [7] . To date , it has proven difficult to relate this genomic variation to differences in parasite biology or host preference . This is because all microsporidia sequenced so far share a broadly conserved core proteome , with differences in genome size due largely to changes in gene density , transposon content , and expansions of uncharacterized , lineage-specific or fast evolving protein families [8]–[10] . As the main differences in coding capacity among sequenced microsporidian lineages , it seems reasonable to hypothesize that these lineage-specific or fast evolving proteins play a role in mediating host-parasite interactions . However , because they lack any detectable similarity to genes from model eukaryotes , the functions of these proteins are difficult to predict using bioinformatics . Combined with the current lack of a system for genetic manipulation in these parasites , this makes understanding the basis of host-microsporidian interactions extremely challenging . Beyond identification of proteins of the spore wall and polar tube , very little known about molecular basis of spore germination and eversion of the polar tube , or the exact details of how the microsporidian sporoplasm is transferred into the host cell . Even though microsporidia can have drastic effects on the organization of the host cell , we know little of the virulence factors and effector proteins that bring about these changes or how they are delivered into the host cell environment either directly or via the parasitophorous vacuole in human infective species such as E . cuniculi and E . intestinalis . Spraguea lophii is a microsporidian that infects the monkfish Lophius piscatorius and Lophius budegassa , inhabiting both the North Atlantic and Mediterranean regions [11] . Compared to those microsporidia that have been sequenced so far , S . lophii is an attractive model for identifying microsporidian effector proteins and investigating host-parasite interactions , despite the current lack of an in vitro culture system [12] . Infection with S . lophii results in the formation of xenomas , large clusters of spore-filled cells in the vagal nerves of the fish that can be several centimeters in diameter and contain spores in various stages of development ( Figure 1 ) . In the monkfish , no fitness effects are known to be associated with Spraguea infection , and the prevalence rate can be as high as 83% [13] , [14] . However , xenoma formation in salmon infected with related microsporidia commonly localizes to the gills and is a considerable threat in aquaculture [15] . S . lophii spores are easily purified from xenomas in large quantities , providing an opportunity to study germination and to perform experiments that are difficult or impossible in other microsporidia due to the difficulty of obtaining sufficient amounts of parasite material . In addition , S . lophii is the first microsporidian parasite of fish to be sequenced , with the potential to provide new insights into host-parasite interactions in these economically important vertebrates . Our aim here is to provide a genomic resource that will facilitate future work on this promising model microsporidian . S . lophii was the first microsporidian to be explored with genome-scale sequencing , and 120 Kb of the genome has previously been published [16] . Here we present 4 . 98 Mb of unique sequence from the S . lophii genome as determined by Illumina sequencing , representing 70–80% of the complete genome ( estimated at 6 . 2–7 . 3 Mb [17]–[19] ) and , based on our analyses , the great majority of the coding DNA . We investigated the evolution of the S . lophii proteome , using OrthoMCL [20] to identify proteins that are unique to S . lophii and may therefore be associated with the unique xenoma formation seen in microsporidian infections of fish . To explore the utility of S . lophii as a model for understanding core features of microsporidian biology , we combined complex mix proteomics with TruSeq transcriptomics to characterize the proteins expressed and secreted during spore germination , a key time point in the lifecycle of this intracellular parasite . Our results highlighted the importance of microsporidia-specific and fast evolving proteins in germination and host interaction .
The sequence data are summarized in Table 1 . Our sequencing resulted in 1392 contigs over 500 bp in length to give a total of 4982 Kb ( This Whole Genome Shotgun project has been deposited at DDBJ/EMBL/GenBank under the accession ATCN00000000 . The version described in this paper is version ATCN01000000 ) . These contigs had a maximum length of 46788 bp and an N50 length of 5923 . The average coverage across these contigs is 70× ( Mode = 29× , Median = 59× ) . The overall GC content is low at 23 . 4% , rising slightly to 25 . 7% in protein coding regions . The karyotype has been investigated both by pulse field electrophoresis and by using an ultrathin multiwire proportional chamber-based detector [17]–[19] . This predicts a variable karyotype between isolates from different geographic regions , but the consensus is that there are 15 chromosomes , of which 10–13 are unique , and the overall genome size is estimated at 6 . 2 to 7 . 3 Mb . In our assembly we identified 2 , 573 predicted open reading frames or fragments of , which made up 52% of our assembly . This gives it an intermediate coding density amongst microsporidia , which is consistent with an emerging pattern for microsporidian genomes where coding density decreases with increasing genome size . For example , Trachipleistophora hominis has 34% coding DNA and an estimated genome size of 8 . 5–11 . 6 Mb , while 86% of the 2 . 9 Mb E . cuniculi genome is made up of coding DNA [5] , [8] . Our assembly covers 70–80% of the S . lophii genome based on size estimates , but is likely enriched for coding regions . To evaluate the completeness of our assembly , we searched the S . lophii genome for the presence of genes involved in a range of core metabolic pathways as described previously [8] , [21] ( Table S1 ) . Our assembly encodes most major metabolic pathways in full , including glycolysis and trehalose metabolism as well as a full complement of transfer RNAs ( tRNAs ) and protein kinases ( Table S1 ) . We identified only a few absences , many of which are also absent in related microsporidia such as T . hominis , lending support to our coverage of the protein-coding component of the genome . As an independent check on the completeness of our assembly , we compared the transcripts from our de novo transcriptome assembly ( see below ) to the genes predicted on the genome . Based on BLAST similarity to genes from other microsporidia , there were only 20 additional S . lophii genes in the transcriptome that did not map to the genome assembly , 5 of which represent transcripts from LTR retrotransposons . Taken together , these analyses suggest that our assembly represents a largely complete sampling of the coding component of the S . lophii genome . To investigate the evolution of gene content in Spraguea , we mapped the taxonomic distribution of microsporidian gene families onto a cladogram ( Figure 2A ) ( derived from a multiprotein phylogeny - see below ) . We built gene families using OrthoMCL [20] on a broad sampling of microsporidian genomes , with Homo sapiens and Saccharomyces cerevisiae as opisthokont outgroups . Our analysis included genomes covering a broad taxonomic spectrum of sequenced microsporidia , including Nematocida parisii Ertm1 [22] , T . hominis [8] , Nosema ceranae [23] , E . cuniculi [5] , and Enterocytozoon bieneusi [24] . The results indicated that 19% of the predicted proteins are shared with all sampled opisthokonts , 1% are specific to sampled fungi , 4% are specific to microsporidia and conserved across the group , 3% are found in clusters of proteins only present in T . hominis and S . lophii . 42% the 2499 analysed S . lophii proteins , do not cluster with proteins from any of the other organisms in our analysis and of these 30% cluster with other S . lophii proteins , indicating that they are part of multiprotein families within the S . lophii genome . Table S2 gives a complete classification of the S . lophii proteome by OrthoMCL analysis . E . cuniculi and other microsporidia have a reduced lipid metabolism repertoire [5] , [24] . A key missing step is the initial reaction of fatty acid synthesis , the carboxylation of acetyl-CoA to malonyl-CoA by acetyl-CoA carboxylase . Homologues of both acetyl-CoA carboxylase and the biotin-[acetyl-CoA-carboxylase] ligase that it depends upon are present in the S . lophii genome as well as in the T . hominis and N . parisii genomes , suggesting that they are capable of performing this reaction . However , as in E . cuniculi , no fatty acid synthase is evident , though both a fatty acid elongase and desaturase are present ( Figure 2B ) . Interestingly , the presence of these additional components in S . lophii was predicted on the basis of comparative liquid chromatography of the lipid composition of E . cuniculi and S . lophii spores which showed a higher level of docosahexaenoic acid , an unsaturated fatty acid , in S . lophii than E . cuniculi [25] . In comparison to E . cuniculi , the S . lophii genome also encodes more enzymes for glycerophospholipid synthesis , allowing for a greater variety of interconversions between different types of phospholipid for membrane integration . In contrast , and despite more complexity in some aspects of fatty acid metabolism , no components of the isoprenoid biosynthesis pathway were present in the S . lophii genome , though these are encoded in the E . cuniculi and N . ceranae genomes [5] , [23] ( Figure 2B ) . These are also absent from our transcriptome data , and from the genomes of T . hominis , N . parisii and E . bieneusi . Taken together , these results suggest that several lineages of microsporidia have independently lost the ability to biosynthesize isoprenoids , a capability that is otherwise conserved across the tree of life [24] . This pathway has been shown to be essential in prokaryotes , and whilst some parasitic Apicomplexa have replaced the classical mevalonate biosynthesis pathway with the alternative MEP pathway [26] , it too is absent from these microsporidian genomes . These data suggest that some microsporidia scavenge sterols from the environment . This pathway has been lost in microsporidia with diverse hosts: E . bieneusi ( mammals , insects ) , T . hominis ( mammals , insects ) , N . parisii ( worms ) and S . lophii ( fish ) , suggesting that there is nothing specific about the host biochemical environment that is driving the loss of this pathway . Key components of the RNAi system are encoded by the S . lophii genome , including a dicer protein , an argonaute protein and fragments of an RNA dependent RNA polymerase . This is consistent with emerging genomic data for microsporidia with larger genomes such as N . ceranae and T . hominis that possess transposable elements [8] , [23] . This suggests that RNAi was present in the common ancestor of microsporidia but has been secondarily lost in highly reduced genomes such as E . cuniculi and E . bieneusi [5] , [8] , [24] . A comparable situation exists in ascomycete fungi , where RNAi has been lost from the compact S . cerevisiae genome but is conserved in other budding yeasts [27] . Relatively few proteins with homology to characterized domains are found in S . lophii but no other microsporidia . There are proteins annotated on the basis of similarity to PFAM domains such as kinases , phosphatases and acetyltransferases , which are difficult to relate to specific functions within the cell . One notable exception is a glutamate-ammonia ligase domain containing protein , which can catalyze the generation of glutamine from glutamate and ammonia [28] . This protein is nested between genes with homologs in other microsporidian genomes ( a DNA binding protein and acetyl-CoA carboxylase ) ( Figure S1 ) and in phylogenetic analysis does not fall into a clade with other fungi , but rather with prokaryotes meaning that it is not clear whether this gene was acquired by lateral transfer or by vertical inheritance . Fish excrete their nitrogenous waste products as ammonia across gills , but glutamate-ammonia ligases are expressed in the brains of fish and other vertebrates to protect from fluctuations in ammonia levels [29] . This protein may have a similar role in protection against ammonia stress in the microsporidian: Whilst the fish is alive the microsporidia may be protected by host ammonia defense mechanisms , however , once the fish dies , microbial degradation of the fish can increase ammonia levels [30] . As the spores of S . lophii are embedded deeply within the nervous tissue of the monkfish , they may have to be liberated after the death of the fish , and this glutamate-ammonia ligase may allow the spores to survive fluctuations in ammonia levels in the decaying fish tissue . Despite the relatively close relationship of T . hominis to S . lophii ( Figure 2A ) , these two species share few genes that are not found in other microsporidian genomes in our comparison , and most of these are uncharacterized , lineage-specific or fast evolving genes with no similarity to genes in other lineages . A handful of genes shared between T . hominis and S . lophii have a function that can be predicted on the basis of homology to characterized proteins from model organisms . These include an arsenite transporting ATPase , which might act as an efflux pump in the cell membrane and a cold shock domain protein , which could allow the cell to survive at lower than optimal temperatures [31] . The largest protein family expansion present in S . lophii is a family of proteins containing LRRs ( Figure 3 ) . Fungal genomes generally encode fewer LRR proteins than their animal relatives [32] though expansion of LRR protein families as pathogenicity factors is known in pathogenic fungi [33] . The genome of E . cuniculi encodes just 9 LRR genes in total [5] , yet in stark contrast , we have found 97 ORFs encoding fragments of LRR proteins in the S . lophii genome , 52 of which appear complete ( that is , they have a predicted start and stop codon ) , and 35 of which appear in our transcriptome data . We used PFAM and MEME [34] , [35] to identify common conserved motifs , SignalP 4 . 1 and TargetP 1 . 1 to look for presence or absence of a secretion signal , and TMHMM 2 . 0 [36] , [37] to look for evidence of transmembrane domains that could anchor the proteins in the membrane of the parasite as a potential LRR receptor protein . MEME analysis shows the presence of three different leucine-enriched motifs in the proteins ( Figure 3 ) . We also found that the majority of the proteins ( 37/52 ) have a predicted signal peptide but no transmembrane domain , meaning that they are potentially a family of secreted parasite effector proteins ( Figure 3 ) . Leucine rich repeat proteins often mediate protein-protein interactions , particularly through the formation of dimers [38] . One possibility is that , if these microsporidian LRR proteins are secreted into the host , they could potentially interfere with the formation of dimers of host proteins , disturbing the functional dimer-monomer cycle by sequestering them into inactive dimers , a mechanism seen in mammalian cells [39] . Perhaps surprisingly , similar sequences are also found in the distantly related microsporidian parasite of humans , Vittaforma corneae , but in no other microsporidian species for which there is an available genome sequence ( Table S3 ) . A large family of leucine-rich proteins was recently reported in another microsporidian , T . hominis , although the two families are not related and no predicted secretion signals were reported for the T . hominis family [8] . Although the S . lophii genome encodes 97 LRR-containing ORFs , we have evidence of expression for only 35 . Therefore , we cannot exclude the possibility that some family members are never expressed and are pseudogenes . In other eukaryotic parasites such as Trypanosoma and Giardia , large protein families contain many pseudogenes which provide the genetic variation for the ongoing process of antigenic switching [40] , [41] . If microsporidian multigene families interact with the host , switching of expression from gene to gene may allow escape from the fish adaptive immune response over the course of infection [42] . The cDNAs of artificially germinated spores were sequenced by the non strand-specific TruSeq approach using Illumina sequencing . This Transcriptome Shotgun Assembly project has been deposited at DDBJ/EMBL/GenBank under the accession GALE00000000 . The version described in this paper is the first version , GALE01000000 . We used Trinity RNA-Seq [43] to assemble the spore transcriptome , and RSEM [44] to quantify the relative abundance of transcripts . Our de novo transcriptome assembly contained 12 , 932 unique transcripts , of which 2 , 896 mapped to the S . lophii genome . Although relatively small in number , these transcripts made up 67 . 7% of the transcriptome by abundance , and likely represent the majority of the S . lophii genes in the dataset . 2 , 514 of these transcripts mapped to existing S . lophii genes , so that 1 , 598 of the predicted 2 , 539 open reading frames had at least one matching transcript . A small proportion of transcripts ( 398 ) mapped to regions of the genome assembly without gene predictions; manual inspection of these cases provided evidence for 30 additional genes that had been missed by the initial annotation process . The remaining third of transcripts that did not map to the S . lophii genome assembly might represent contaminants , unmapped S . lophii genes , or artifacts of the assembly process . To distinguish between these possibilities , we searched these transcripts against the NCBI nr database using BLASTX [45] . Many showed high sequence identity to Pseudomonas and Flavobacterium genes and likely represent contaminants . We did , however , identify 20 additional genes with best BLAST hits among the Microsporidia , particularly T . hominis and Vavraia culicis; these are most likely S . lophii genes from the unassembled portion of the genome ( Table S4 ) . Interestingly , this set of 20 genes included one LTR and four non-LTR retrotransposons with similarity to those found on the T . hominis genome [8] . While the T . hominis retrotransposons are largely fragmented and pseudogenized , these results demonstrate that at least some of their homologues in S . lophii remain active . Overall , our analyses of the S . lophii transcriptome provided support for the completeness of our genome assembly and for our gene calling approach , and also provided some new insights into microsporidian biology . Of the 1 , 986 genes in our genomic data that have complete open reading frames , that is , they have a start and a stop codon , 265 have complete coverage and for these , the RNA transcript was on average 132 base pairs longer than the gene . Five of these show more than one gene in the transcripts . However , given the short read-length of Illumina sequences and the possibility that transcripts of adjacent divergent genes could erroneously assemble , it is not possible to say whether these transcripts are overrunning into downstream genes as seen in other microsporidian species [46] , [47] . Interestingly , although the most abundant transcripts corresponded to 18S ribosomal RNA ( Table 2 ) , the most highly expressed protein in S . lophii spores is an uncharacterized ORF , with homologues found only in a limited number of other microsporidia . Indeed , while the list of highly transcribed proteins contained many of the expected candidates ( ribosomal proteins , ATP-binding proteins , transcription factors , and proteins involved in energy metabolism ) , these were intermingled with a number of uncharacterized proteins annotated as hypotheticals ( see Table 2 ) , strongly suggesting an important role for novel , lineage-specific or very fast evolving proteins in S . lophii and microsporidian biology . An interesting observation is that microsporidia-specific proteins conserved in multiple species make up a larger proportion of the transcriptome than Spraguea-specific proteins ( Figure 4 ) , and a larger proportion of the transcriptomic data than they do of the genomic data . These potentially novel proteins originating in the common ancestor of microsporidia may be good targets for future experimental work on the maintenance of the parasitic lifecycle both in S . lophii and other microsporidian species . Our de novo transcriptome assembly also enabled us to investigate the splicing of putative introns in S . lophii protein-coding genes . The number of introns in microsporidian genomes is greatly reduced compared to their opisthokont relatives [48] . E . cuniculi , N . ceranae , and T . hominis encode a relatively small ( 6–78 ) number of spliceosomal introns , which are largely confined to ribosomal proteins , while the Nematocida genus appears to have lost both introns and the splicing machinery entirely [22] . S . lophii does encode conserved components of the splicing machinery , so we searched its coding sequences for introns using a two-step approach . First , we scanned the genome with a consensus microsporidian intron motif built from comparisons of the introns in E . cuniculi and T . hominis [5] , [8] , [49] . This search returned hits to 8 genes , 6 of which encode ribosomal proteins ( Figure 5 ) ; the two non-ribosomal proteins included genes encoding the DNA replication licensing factor Mcm1 and a poly ( A ) binding protein . We then searched the S . lophii transcriptome for transcripts containing deletions relative to the genome assembly , which might also indicate the presence of introns . Surprisingly , this analysis identified only two transcripts from which the predicted intron sequence had been spliced , corresponding to two of the eight genes identified by our motif scan ( ribosomal protein S23 and poly ( A ) binding protein , Figure 5 ) , transcripts for the other six genes still contained the putative intron motif . A comparison of the sequences of the two actively spliced introns with those of the other intron-like sequences revealed three striking differences ( Figure 5 ) : the spliced introns are much longer , are out of frame with the coding sequence , and are located further downstream from the 5′ end of the gene ( 89 and 152 nucleotides 3′ of the start codon , as opposed to directly adjacent to that codon in all other cases ) . Of the eight putative intron-containing genes we identified in Spraguea , five have orthologues in E . cuniculi that also contain an intron ( S17 , L27a , S24 , L5 and poly ( A ) binding protein ) , and the efficiency with which those introns are spliced parallels our results with the S . lophii transcriptome . The introns in E . cuniculi S17 , L27a , S24 and L5 , for which we did not detect splicing in Spraguea , are also short [49] and are among the least efficiently spliced genes in E . cuniculi , with less than 15% of transcripts experiencing splicing ( a figure which drops to 5% for L5 ) [50] . In contrast , the E . cuniculi orthologue of the actively-spliced poly ( A ) binding protein contains the longest and most frequently spliced intron in E . cuniculi , with over 80% of transcripts spliced . Thus , it appears that the properties determining intron splicing efficiency are conserved between these two distantly related microsporidia; it will be interesting to see if they hold more generally for other intron-containing microsporidian genomes . These observations raise the question of how genes containing intron-like sequences that are rarely , if ever , spliced can be adequately expressed in Spraguea . The genes containing these motifs encode some of the most widely conserved and functionally important proteins in cellular life forms , including components of the ribosome and a DNA replication factor . Several of these intron-containing proteins were identified in our whole cell protein analysis of germinated and non-germinated spores . Ribosomal proteins S27 , S24 and L5 were present in our germinated sample and S27 and L5 were also present in the proteome of dormant spores ( Table S5 ) . For two of the six genes ( S24 and L5 ) , inspection of the intron-like sequence suggests a simple explanation: indels in these sequences have caused a frameshift such that the intron can be read through from the upstream ATG without encountering an in-frame stop codon . Thus , a full-length protein containing a short N-terminal insertion could be expressed from these transcripts in the absence of splicing . The other four introns contain in-frame stop codons such that translation from the upstream ATG is not possible unless the intron is spliced . We considered the possibility that translation could instead begin at an alternative start codon downstream of the intron . However , initiation at the next available , in-frame ATG would result in substantial N-terminal deletions ( covering 25–37% of the coding sequence ) for three of these genes , and in the case of ribosomal protein S17 no suitable alternative start codon is available . Thus , it remains unclear whether these genes can be expressed without splicing , or whether translation depends on a rate of splicing too low to be detected in our assay; it was recently suggested that low rates of transcript degradation might partially ameliorate this problem in E . cuniculi [50] . Dissecting the interactions between microsporidia and their host requires an understanding of the process of spore germination , in which the spore leaves dormancy and rapidly expels a long polar tube , through which the spore's cellular contents exit the spore and enter the host cell . The specific host cell stimuli inducing microsporidian spore germination are unidentified and likely complex , however several successful methods for artificial spore germination in-vitro have been described in a range of species [51]–[53] . At present , both the changes within the spore that trigger germination , and the identity of the secreted effector proteins used by the parasite gain entry into the host cell and control its biology are unknown . Genetic manipulation is a powerful tool for identifying these factors in many pathogens , but is not yet available for any microsporidian . However , S . lophii xenomas are densely packed with spores , providing an abundant source of parasite material for proteomic comparisons between the dormant and germinated spore stages . To identify any proteins present in artificially germinated but not dormant spores , we analyzed whole protein extractions of both lifecycle stages with mass spectrometry of complex protein mixtures . After pooling and filtering 3 biological replicates our analysis showed no consistent variation at the proteomic level between the two lifecycle stages ( Table S5 ) . We did find components of many core pathways in germinated and non-germinated spores , such as histones , heat shock protein and ribosomal proteins . We also find glycolytic enzymes in both germinated and non-germinated spores , which is consistent with recent work that glycolytic pathways are specifically active in the spore stage [8] , [54] . Two components of the secretory pathway ( Sec23 and Sec24 ) were identified only in germinated spores , which may indicate its activation specifically upon germination , however these were not consistently found in all three replicates and overall we found a surprisingly conserved repertoire of proteins between the two samples . It may be that germination happens too rapidly to allow for translation , with microsporidia pre-packaging the proteins needed for immediate use upon recognition of the germination stimulus and therefore that obvious changes in protein complement may come later in development in meront and sporogonial stages . Alternatively it may be that the samples are dominated by highly expressed housekeeping proteins , and the proteins that vary between the two samples may be present at low levels not easily detectable by complex mix proteomics . Next , we investigated the complement of proteins present in the extracellular medium after in-vitro germination . Here , we consistently retrieved a small subset of proteins that were visualized by SDS-PAGE ( Figure S2 ) . Importantly , no proteins were identified from the supernatant of non-germinated S . lophii spores , suggesting that the identified proteins are released specifically by the parasite upon germination . These could be proteins released by the sporoplasms on early infection or by the spore during germination . We retrieved 37 proteins from three quality-filtered replicates ( Figure 6 ) . Of these , 11/37 proteins are predicted by SignalP 4 . 1 to have a secretion signal and 17 are predicted by TargetP 1 . 1 to be directed to the secretory pathway; in some cases these predictions overlap ( Figure 6 ) . Proteins secreted to the extracellular medium range in size from 101 amino acids to 935 amino acids however there is no obvious correlation between size and the presence or absence of a predicted secretion signal ( Figure 6 ) . Five proteins are consistently present in all three replicates . Three of these have no similarity to any other proteins in the NCBI nr database ( e<1×10−5 ) ( SLOPH 477 , 723 , 762 ) , while two of the proteins are found in other microsporidia ( SLOPH 1766 , 1854 ) . One of these has features which make it particularly interesting in the context of potential effector proteins: it is part of a multigene family whose members have predicted Ricin-B-lectin domains and is found in several copies in S . lophii , as well as in several other microsporidian genomes ( SLOPH 1766-see below ) . The other secreted protein shared with other microsporidia has significant sequence similarity to a spore wall protein ( SWP7 ) in Nosema bombycis ( SLOPH 1854 ) . Ten proteins are found in two of three replicates . Six of these proteins have no similarity to any proteins in the NCBI database ( e<1×10−5 ) and are potentially unique to S . lophii , demonstrating that species-specific innovations may play a crucial role in S . lophii invasion of the host cell . Four proteins are found in S . lophii and other microsporidia , including SLOPH 1749 , which contains an exonuclease/endonuclease/phosphatase domain , SLOPH 2344 , a spore wall protein and another RICIN-B lectin domain protein ( SLOPH 691 - see below ) . A chitin deacetylase , which is conserved across the Microsporidia , was also identified in 2/3 biological replicates . The orthologue of this protein has been studied in E . cuniculi . In this species it is expressed at high levels during sporogonial lifecycle stages and localizes to the endospore , where it accumulates in paramural bodies [55] , [56] . However , later functional analysis demonstrated that the protein was unable deacetylate chitooligosaccharides or bind chitin or any glycans , suggesting a non-canonical function [57] . Urch et al speculate that this chitin deacetylase may have evolved from a carbohydrate active enzyme to a lectin and may bind carbohydrates in the E . cuniculi cell wall [57] . A total of five microsporidia-specific uncharacterized proteins were observed in at least 2/3 biological replicates ( SLOPH 1854 , 1766 , 2344 , 1749 , 691 ) ( Figure 6 ) . Orthologues of these proteins display a varied taxonomic distribution and secretion prediction profile across microsporidia , although two of them - SLOPH 2344 and SLOPH 1749 - are present in all microsporidia examined ( with the exception of N . parisii ) and are predicted to be secreted in all species expect E . intestinalis and E . bieneusi ( SLOPH 2344 ) , or just E . bieneusi ( SLOPH 1749 ) ( Table S6 ) . Some proteins lacking predicted secretion signals are potentially highly expressed proteins that are released during the germination protocol and are therefore potentially false positives , for example histone proteins and translation elongation factor which were found in single replicate . Others , that are present in 2 or 3 replicates , may be secreted via a non-canonical secretion signal not detected by bioinformatics prediction programs . The S . lophii-specific proteins may be host-driven innovations that mediate interactions specifically with the fish host , representing cases of lineage-specific adaptation or microsporidian-specific proteins that are fast evolving and not easily recognized between species . Some of the identified proteins are part of multigene families with other homologs in the S . lophii genome ( Starred in Figure 6 , Table S7 ) though proteomic data correspond to one member of the family . In agreement with our transcriptome analysis , these results demonstrate the importance of uncharacterized , hypothetical proteins in microsporidian germination , host cell invasion and early infection . This approach also represents a powerful method for identifying the presumably small proportion of novel parasite effectors or virulence factors from among the hundreds , or thousands , of hypothetical proteins found in sequenced microsporidian genomes . The need for such streamlining approaches is obvious , as a genetic manipulation protocol for microsporidia is currently lacking and hence protein characterization relies on heterologous expression systems or indirect assays which are time consuming and not easily applied to a large number of proteins . Two of the identified secreted proteins ( SLOPH 1766 and SLOPH 691 ) are conserved in other microsporidia and show similarity to RICIN B-lectin proteins , with weakly conserved motifs involved in carbohydrate binding . Lectins have diverse roles in parasites , and can mediate adhesion of the parasite to the host cell during infection , but they can also play roles in immune evasion and their binding to host proteins can trigger different developmental pathways in the parasite [58] , [59] . The members of this family form clusters in the genomes of E . cuniculi and N . ceranae ( Figure 7A ) , reminiscent of clusters of effector proteins in fungal pathogens such as Ustilago maydis [60] and the RXLR and Crinkler effector families in oomycetes of the genus Phytophthora [61] . In E . cuniculi , four proteins are found in a single syntenic block , whereas in N . ceranae , a block of six proteins sit together in the genome with two others elsewhere . Interrogating the E . cuniculi genome , which has well-assembled chromosomes , these genes are not found at the end of chromosomes in the subtelomeric regions as is the case for effector proteins in other parasite species [62] . It is difficult to draw conclusions about the relative genomic location of the corresponding genes from our S . lophii data as several are found on short contigs , but we did identify two clusters of two genes ( Figure 7A ) . A phylogeny of this protein family suggests that most of the paralogs within each species arose from species-specific duplications ( Figure S3 ) , again mirroring the pattern seen for effector families in other eukaryotic pathogens such as Ustilago maydis [60] , Phytophthora [61] and Trichomonas [63] . The distribution of lectin-like proteins and expanded gene families were plotted onto a multigene phylogeny of microsporidia to give an overview of their distribution across the phylum ( Figure 7B ) ; the patchy distribution of these families suggests a process of differential loss and expansion during the radiation of microsporidia . In particular , we did not detect homologues of the lectin-like proteins in the N . parisii genome , suggesting that this family may have evolved after the divergence of N . parisii from the other sequenced microsporidia .
Although the S . lophii genome is larger than that of E . cuniculi , there is no clear trend towards a greater retention of broader ancestral metabolism . Like T . hominis , S . lophii has apparently lost the ability to biosynthesize isoprenoids , although it has retained more enzymes involved in other aspects of lipid metabolism . These results demonstrate that the loss of major biosynthetic pathways did not occur only in the common ancestor of all microsporidia , but has been an ongoing process throughout the evolution of the group , giving rise to important lineage-specific differences in metabolism . Beyond reduction in genome size , a relative scarcity of introns is a conserved and striking feature of microsporidian genomes . The remaining introns tend to be short and , in many cases , located at the 5′ ends of genes encoding ribosomal proteins [5] . Our comparisons of splicing in S . lophii and E . cuniculi suggest that the molecular determinants of splicing efficiency are conserved in these distantly-related species , with longer , out-of-frame introns experiencing the highest levels of splicing . The observation that some inefficiently spliced , intron-like sequences can be read through without producing a truncated or frameshifted protein suggests a plausible , though speculative , mechanism by which introns could be lost from microsporidian genomes: once read-through is possible , subsequent deletions could reduce or remove the newly expressed insertion , resulting in complete loss of the former intronic sequence . We note , however , that the introns we compared in our assay are conserved between S . lophii and E . cuniculi , and so have not been lost in the period of time since the divergence of these species . One of the most striking features of the S . lophii genome is the presence of a leucine rich repeat protein family , which may represent an expanded gene family of pathogenicity factors . Such expanded gene families are characteristic of other fungal pathogens such as Batrachochytrium dendrobatidis and Blumeria graminis , have also been reported in the microsporidia T . hominis , E . cuniculi , Anncaliia algerae and Vittaforma corneae [8] , [64]–[66] , and are also considered of importance in host-parasite interactions more generally [67] . Whilst the S . lophii members of this family have predicted N-terminal secretion signals , there is no evidence of an obvious conserved peptide motif involved in directing protein secretion into the host such as the oomycete crinkler motifs or the conserved tripeptide motif found in B . graminis which could help to define the microsporidian secretome [65] , [68] . In addition to the conserved microsporidian proteome and an expanded complement of LRR proteins , S . lophii encodes a large number of predicted hypothetical proteins . Some of these were only expressed at low levels in our transcriptome analysis , potentially representing false positive ORF calls , but there were also a significant number of highly expressed microsporidia-specific transcripts and together with S . lophii-specific transcripts , these made up 39 . 1% of the total expression in S . lophii spores ( Figure 4 ) . Thus , lineage-specific or fast evolving proteins likely play an important role in S . lophii biology , and the same may well be true for all microsporidia . Although we now have genome sequences for a number of microsporidia , our understanding of their basic biology , pathogenicity and host interactions is limited by the current lack of a genetic manipulation system for these intracellular parasites . Recent work has shown that identification of proteins bearing secretion signals through bioinformatics can reveal interesting strategies by which microsporidia may alter the host environment in their favor . Cuomo et al . have identified functional secretion signals in a family of microsporidian hexokinases; when secreted , these effectors may stimulate host metabolism , providing more metabolites for the parasite [22] . These results demonstrate the utility of bioinformatics in predicting signal peptides and identifying microsporidian proteins that may be targeted to the cell surface or secreted . However , signal peptide prediction tools are trained on model eukaryotes , and their accuracy in predicting the full secretome from the highly divergent sequences of microsporidia is unclear . Even when these signals are accurately identified , heterologous characterization is laborious and expensive . One way forward may be through alternatives to genetic transformation , such as those we have explored in this article . Here we have used secretion proteomics to detect candidate virulence factors that are likely to be secreted from the parasite as it germinates , leaves the polar tube and enters the host cell . Two of these identified secreted proteins are a part of a family that is broadly conserved across the Microsporidia , with local duplications giving rise to syntenic blocks of family members in several microsporidian genomes . Based on their sequence similarity to lectin domains , they may be involved in binding to carbohydrates on host proteins , and their conservation suggests they play an important general role in microsporidian parasitism . We combined our secretion proteomics with an analysis of the S . lophii transcriptome , which identified a further set of highly expressed , microsporidia-specific hypotheticals . These proteins are ideal candidates for further characterization to better understand the molecular basis of parasitism in the Microsporidia , both within the parasite cell and in its host interactions .
DNA was extracted from two separate clusters of cysts both collected from local Monkfish ( Lophius piscatorius ) landed at either Plymouth or Brixham caught in the North Atlantic . Spores were separated from fish material in a three-step process . Xenomas filled with microsporidia were removed from the fish tissue manually . Samples were then soaked overnight at 4°C in EDTA ( 10 mM ) containing Triton X 100 at 0 . 05% and trypsin 0 . 025% ( w/v ) . The samples were then homogenized in a glass homogenizer until they formed a fine suspension , and washed three times in sterile PBS . The spore suspension was then cleaned by centrifugation through 100% Percoll ( Sigma ) at 4°C at 1600×g for 15 minutes . Samples were further washed three times in sterile 1× PBS before resuspension and storage at 4°C in 1 ml of sterile 1× PBS with addition of an antibiotic cocktail of 10 µg/ml Ampicillin , Penicillin/Streptomycin and Kanamycin . Aliquots of 200 µl of purified spores were resuspended in 400 µl TE 10/1 pH 7 . 5 and ground in a pestle and mortar in liquid nitrogen for 15 minutes . Powdered frozen material was transferred to 800 µl phenol ( pH 7 . 9 ) and mixed by inversion then centrifuged at 10 , 000×g for 10 minutes . 400 µl chloroform was added to the aqueous supernatant and this was centrifuged at 10 , 000×g for 5 minutes . DNA was precipitated from the aqueous layer using a standard ethanol precipitation protocol [69] . For sample one , the DNA was sheared using a Covalis Nebuliser . Sheared fragments of approximately 400 bp in size were selected using a polyacrylamide gel and used to create a library using the Illumina paired-end sample preparation protocol ( revision A June 2008 ) . A single lane of paired-end 36 bp sequence reads was sequenced . For sample two , DNA was sheared by Biorupter sonication and fragments of approximately 600 bp were selected by polyacrylamide electrophoresis for library preparation as above . A single lane of paired-end reads of 76 bp was sequenced giving a total of 325 Mb of data . The data from the two runs were pooled and assembled together using Velvet 0 . 7 . 50 using a kmer length of 31 and coverage cutoff of 5 . Estimated expected kmer coverage using automatic calibration was 6 . Contigs of 500 bp or larger were retained and analysed further . This resulted in a set of 3080 contigs , which showed a bimodal distribution of GC content with peaks at approximately 23% GC and 58% GC , indicative of DNA from a second organism ( based on sequence identity , a close relative or strain of Pseudomonas fluorescens ) . A histogram of GC content revealed two distinct distributions with little overlap ( Figure S4 ) ; we therefore discarded all contigs with GC content greater than 40% as potential Pseudomonas contaminants . All retained contigs were used to BlastN search gi229587578 Pseudomonas fluorescens SBW25 chromosome and gi146343893 Pseudomonas fluorescens SBW25 plasmid pQBR103 , to verify that there was no P . fluorescens DNA in the remaining contigs . Removed high GC contigs were also used to BLAST search the same genomes and any contigs with a BLASTN hit value higher than 1×10−30 were used to BLASTX search the same microsporidia used in our comparative analysis to verify that there was no obvious microsporidian content . Remaining contigs were parsed to find unknown stretches of DNA left by paired-end assembly of 5 base pairs or more and split at these points . This left a set of 1392 contigs that were further analyzed . These contigs were annotated using Artemis 13 . 2 . 0 . All ORFs of 100 amino acids or more were analyzed by BLASTP and PFAM search and an annotation given on the basis of these searches [35] . Each contig was searched for the presence of tRNAs using tRNAscan-SE v . 1 . 23 [70] , which were annotated onto the contigs using Artemis . 56274 proteins sequences from eight species ( predicted proteomes of H . sapiens , S . cerevisiae , N . ceranae , E . bieneusi , E . cuniculi , N . parisii , T . hominis and those 2499 S . lophii proteins without predicted frameshifts ) were clustered using OrthoMCL: BLAST e-value cutoff = 1e-5 , Inflation value = 1 . 5 . Output files were parsed and sorted into categories shown in Figure 2A . 300 µl of purified spores were washed 3 times in 1× PBS . A non-germinated control sample was then treated with 100 µl of 0 . 25 mM EGTA per 100 µg of spores to inhibit germination . The sample to be germinated was treated with 100 µl of 0 . 5 M Gly-Gly buffer ( pH 7 . 0 ) per 100 µg of spores and incubated at room temperature for 30 minutes . 40 µl of Calcium Ionophore A23187 ( 1 µg/µl ) ( Sigma-Aldrich ) dissolved in DMSO was added to the sample followed immediately by 40 µl of 0 . 5 M Gly-Gly buffer ( pH 9 . 0 ) per 100 µg of spores . Germination was verified instantly by light microscopy and efficiency was estimated to reach a maximum of 80% . 200 µl of germinated S . lophii spores were frozen with liquid nitrogen and disrupted manually using a pestle and mortar . RNA was then extracted following the standard TRIzol ( Invitrogen ) protocol . RNA was resuspended in 50 µl Milli-Q water ( Millipore ) and quantified as 2260 ng/ul ( absorbance 260/280 = 2 . 12; 260/230 = 2 . 41 ) . RNA integrity was verified on 1 . 5% TAE agarose gel prior to sequencing . PolyA RNA was isolated from 4 µg total RNA and the TruSeq library was prepared according to the Illumina TruSeq RNA sample preparation guide ( Part # 15008136 Rev . A November 2010 ) . 15 cycles of PCR were used to amplify the library and 5 µl of the 6 nM library was loaded onto a flow cell with two other TruSeq libraries and run on the HiSeq 2000 with a paired-end 100 bp run . The raw Illumina reads obtained from RNA sequencing were filtered with fastq-mcf [71] to remove adapter sequences and low quality reads ( with a quality score <28 ) . The filtered reads were assembled using the Trinity package [43] . Reads were mapped back onto the transcriptome assembly using Bowtie [72] , and the abundance of each transcript was estimated using RSEM [44] . 500 µl of purified S . lophii spores were germinated using the protocol described previously . Germinated S . lophii spores along with non-germinated control were then spun at 12 , 000×g for 15 minutes at 4°C . The resultant supernatant was then collected and concentrated using Millipore Amicon 3 kDa centrifugal concentration column at 2 , 000×g for 30 minutes at 4°C . The concentrated extracellular protein was then quantified and checked on 12% SDS gels . Complex mixtures of extracellular secreted proteins from germinated and non-germinated samples were sent for analysis on an 6520 accurate mass quadrupole time of flight ( Q-TOF ) mass spectrometer ( Agilent Technologies ) and resultant masses were searched against our translated S . lophii genome database . The experiment was conducted in triplicate and any protein without two distinct peptide hits and a percent score peak intensity ( % SPI ) of ≥60% was removed from the analysis . | Microsporidia are unusual intracellular parasites that infect a broad range of animal cells . In comparison to their fungal relatives , microsporidian genomes have shrunk during evolution , encoding as few as 2000 proteins . This minimal molecular repertoire makes them a reduced model system for understanding host-parasite interactions . A number of microsporidian genomes have now been sequenced , but the lack of a system for genetic manipulation makes it difficult to translate these data into a better understanding of microsporidian biology . Here we present a deep sequencing project of Spraguea lophii , a fish-infecting microsporidian that is abundantly available from environmental samples . We use our sequence data combined with germination protocols and complex-mix proteomics to identify proteins released by the cell at the earliest stage of germination , representing potential pathogenicity factors . We profile the RNA expression pattern of germinating cells and identify a set of highly transcribed hypothetical genes . Our study provides new insight into the importance of uncharacterized , lineage-specific and/or fast evolving proteins in microsporidia and provides new leads for the investigation of virulence factors in these enigmatic parasites . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"parastic",
"protozoans",
"mycology",
"genomics",
"protozoology",
"host-pathogen",
"interaction",
"microbiology",
"biology",
"proteomics",
"parasitology"
] | 2013 | The Genome of Spraguea lophii and the Basis of Host-Microsporidian Interactions |
Inorganic arsenic is a carcinogen , and its ingestion through foods such as rice presents a significant risk to human health . Plants chemically reduce arsenate to arsenite . Using genome-wide association ( GWA ) mapping of loci controlling natural variation in arsenic accumulation in Arabidopsis thaliana allowed us to identify the arsenate reductase required for this reduction , which we named High Arsenic Content 1 ( HAC1 ) . Complementation verified the identity of HAC1 , and expression in Escherichia coli lacking a functional arsenate reductase confirmed the arsenate reductase activity of HAC1 . The HAC1 protein accumulates in the epidermis , the outer cell layer of the root , and also in the pericycle cells surrounding the central vascular tissue . Plants lacking HAC1 lose their ability to efflux arsenite from roots , leading to both increased transport of arsenic into the central vascular tissue and on into the shoot . HAC1 therefore functions to reduce arsenate to arsenite in the outer cell layer of the root , facilitating efflux of arsenic as arsenite back into the soil to limit both its accumulation in the root and transport to the shoot . Arsenate reduction by HAC1 in the pericycle may play a role in limiting arsenic loading into the xylem . Loss of HAC1-encoded arsenic reduction leads to a significant increase in arsenic accumulation in shoots , causing an increased sensitivity to arsenate toxicity . We also confirmed the previous observation that the ACR2 arsenate reductase in A . thaliana plays no detectable role in arsenic metabolism . Furthermore , ACR2 does not interact epistatically with HAC1 , since arsenic metabolism in the acr2 hac1 double mutant is disrupted in an identical manner to that described for the hac1 single mutant . Our identification of HAC1 and its associated natural variation provides an important new resource for the development of low arsenic-containing food such as rice .
Inorganic arsenic is a non-threshold class-1 chronic exposure human carcinogen [1] , and its elevated level in rice ( Oryza sativa ) produced in Bangladesh , China , and India is known to pose a significantly elevated cancer risk in these populations , which eat rice at the high levels typical of many Southeast Asian countries [2] , [3] . Products derived from rice ( such as baby food ) and juices ( such as apple and grape ) can also contain inorganic arsenic at levels that pose a health risk . Several brands of baby food and juice contain arsenic concentrations that exceed the United States federal arsenic limit for drinking water [4] , [5] raising significant health concerns in the US and Europe [6]–[8] . Because of this serious and widespread food safety concern research into understanding the mechanisms driving arsenic accumulation in plants has become a priority [9] . Arsenate is the most prevalent form of arsenic in the environment and its similarity to phosphate allows it to be taken up by plants via the phosphate uptake transporters [10] . On exposure to arsenate , plants rapidly respond by suppressing expression of the PHT1;1 gene encoding an arsenate/phosphate transporter , and by removing the transport protein from the plasma membrane [11] to limit arsenate uptake . Expression of PHT1;1 in response to arsenate is modulated by the transcription factor WRKY6 [11] . Though this response helps limit arsenate uptake , it does not eliminate it , and the first step after plants take up arsenate is its chemical reduction to arsenite [12] . In the arsenite form arsenic is either extruded back out of roots [13] , [14] , transported to the shoot ( and on to the grain ) [15] , [16] , or detoxified by complexation by thiol groups in phytochelatins and compartmentalised as a complex into the vacuole [12] , [17]–[19] . The molecular components that drive these processes downstream of arsenate's conversion to arsenite are starting to be understood [20] . Using sequence homology with the known Saccharomyces cerevisiae arsenate reductase ACR2 [21] , or functional complementation of a yeast mutant lacking a functional ACR2 , genes encoding ACR2-like enzymes have been isolated from the plants Arabidopsis thaliana , rice , Pteris vittata , and Holcus lanatus [22]–[25] . Initial evidence using RNA interference ( RNAi ) suggested that suppression of the ACR2-like gene in A . thaliana caused a significant increase in sensitivity to arsenate and increased accumulation of arsenic [23] . However , more recent experiments using two independent T-DNA insertion alleles of the ACR2-like gene ( acr2-1 and acr2-2 ) showed that specific loss-of-function of this gene in A . thaliana has no observable impact on arsenate tolerance , the accumulation of arsenate or arsenite , or the efflux of arsenite from roots [26] . The function of this ACR2-like gene in arsenic metabolism in A . thaliana now appears unlikely . Natural genetic variation is a powerful resource for investigating the molecular function of genes [27] . A . thaliana is broadly distributed throughout the northern hemisphere , and its genome contains extensive diversity associated with broad phenotypic variability [28] and local adaptation [29]–[34] . This natural variation has been used to identify specific genes involved in controlling variation in many traits [28] . Connecting natural genetic variation with its associated phenotype ( s ) has traditionally been achieved using populations of recombinant inbred lines ( RILs ) in which homozygous alternative alleles are segregating . Such populations have high sensitivity to detect causal loci . However , they have low resolving power because of the limited number of recombination events , making identification of causal genes more difficult . Furthermore , because each mapping population is usually generated from a cross between two parental accessions , only a limited amount of natural allelic diversity is captured in these populations , limiting the detection of important minor alleles . Genome-wide association ( GWA ) mapping is an alternative approach to using synthetic RILs , which takes advantage of the large number of historic recombination events within a population . By coupling these events with linked DNA polymorphisms genome-wide , the phenotypic effect of multiple alleles across the genome can be tested . However , unlike in synthetic RILs made from a bi-parental cross , the low frequency of rare alleles in a natural population makes it difficult to detect their phenotypic effect using GWA mapping . Nonetheless , GWA mapping has been successfully used in plants [35] , including A . thaliana [36]–[46] , rice [47]–[52] , and maize [53]–[58] , for the identification of quantitative trait loci ( QTL ) and candidate genes for various ecological and agricultural traits . Here , we report the use of GWA mapping to identify a major locus involved in controlling variation in arsenic accumulation in plants . This locus encodes an arsenate reductase enzyme that lacks the conserved active site of the canonical yeast ACR2 [21] . This arsenate reductase functions to chemically reduce arsenate to arsenite in the outer cell layer of the root to allow efflux of arsenite from the root back into the soil . The localisation of this enzyme in the pericycle also suggests it provides the arsenate reductase capacity to block the transport of arsenic to shoots as the easily transportable phosphate analogue arsenate by reducing arsenate to arsenite in the stele .
In this study we used 349 genetically diverse A . thaliana accessions collected from habitats across the species range [44] , [46] to identify genetic loci controlling leaf accumulation of arsenic . To achieve this , plants were grown in artificial soil containing arsenate at a non-toxic , environmentally relevant , concentration of 0 . 1 µmoles g−1 dry weight ( 7 . 5 mg kg−1 ) . After 5 wk of growth , and prior to flowering , one to two leaves were harvested from each plant and their total arsenic concentration determined by inductively coupled plasma–mass spectrometry ( ICP-MS ) . Leaf arsenic concentrations were found to range from 0 . 15 to 3 . 49 µg g−1 dry weight ( DW ) ( Figure 1A ) . To associate variation in leaf arsenic concentration with variation at specific genetic loci , we performed a GWA analysis . To achieve this , we used the leaf arsenic concentration in 337 of the phenotyped accessions that had previously been genotyped at approximately 248 , 584 genome-wide diallelic single nucleotide polymorphisms ( SNPs ) [46] . Applying a mixed model approach to associate genotype with phenotype at single loci while correcting for population structure [44] , [46] revealed a 47 kb interval on chromosome 2 containing 30 SNPs that are highly associated with variation in leaf arsenic ( p-values<10−5 ) ( Figure 2A , 2B ) , 14 of which were associated with p-values<10−8 . The most highly associated SNP ( −log p-value = 11 . 9 ) at Chr2:9008060 ( TAIR9 genome release ) explained 17 . 6% of the total variation in leaf arsenic concentration across the experiment . No SNPs in other chromosomal regions explaining more than 6% of the variation in leaf arsenic were observed , suggesting the causal gene in linkage with SNP Chr2:9008060 is the major genetic locus controlling natural variation in leaf arsenic accumulation when A . thaliana is grown in soil containing a subtoxic concentration of arsenate . Accessions with a cytosine ( C ) at Chr2:9008060 have average leaf arsenic concentrations 37 . 4% higher than accessions with a thymine ( T ) , and the minor T allele is present in 24 . 3% of the 337 accessions studied . To validate the existence of this leaf arsenic QTL identified by GWA analysis we created a synthetic F2 recombinant population in which alternative alleles of the dialleic SNP Chr2:9008060 were segregating . This population was created by crossing the high leaf arsenic accession Krefeld ( Kr-0 , CS28419 ) , with a C at Chr2:9008060 , to Col-0 that contains average leaf arsenic and has a T at Chr2:9008060 . F1 plants from this cross had leaf arsenic concentrations equal to Col-0 ( Figure 1B ) , indicating that the Kr-0 allele for high leaf arsenic is recessive relative to Col-0 . Mapping the high arsenic locus was performed using extreme array mapping ( XAM ) [44] , [59] , [60] . Arsenic concentrations were measured in leaves from 315 F2 plants ( Figure 1C ) from the Kr-0×Col-0 cross and approximately one-quarter of these plants ( 75 of 315 , X2 = 0 . 24<X20 . 05 ( 1 ) = 3 . 84 ) had high leaf arsenic , suggesting that high leaf arsenic in Kr-0 is controlled by a single major locus . For XAM , 59 of these F2 plants with high leaf arsenic ( As >7 µg g−1 DW ) and 61 plants with low arsenic ( As <0 . 8 µg g−1 DW ) were pooled separately , genomic DNA isolated from each pool and genotyped using the Affymetrix SNP-tiling array Atsnptile 1 . Allele frequency differences between the two pools for all SNPs polymorphic between Kr-0 and Col-0 were assessed [59] and used to determine that the causal locus for high leaf arsenic is located between 8 . 5 and 9 . 5 Mb on chromosome 2 ( Figure 2C ) . This QTL was named High Arsenic Content 1 ( HAC1 ) . Fine mapping based on the genotype and leaf arsenic concentrations of informative recombinants from a new set of 1 , 321 F2 plants from the Kr-0×Col-0 cross determined the causal locus for HAC1 to be within a 40 . 9 kb region between markers CS9MB ( SNP Chr2:8993958 ) and CS9040K ( SNP Chr2:9034914 ) ( Figure 2D , 2E ) . Linkage and GWA mapping located HAC1 to the same region of chromosome 2 containing 18 genes ( Figure 2A , 2E ) . To determine which gene is casual for HAC1 we analysed the arsenic concentration of leaves from 48 T-DNA insertion alleles of genes in this region ( Table S1 ) . From this screen of T-DNA insertion alleles , we identified lines GABI_868F11 and SM_3_38332 with leaf arsenic concentrations similar to Kr-0 , and we named them hac1-1 and hac1-2 ( Figure 2F ) . Both mutants contained T-DNA insertions in gene At2g21045 , suggesting this gene is causal for HAC1 . The distance between the peak SNP from the GWA analysis and At2g21045 is 19 . 8 kb . Sequencing of At2g21045 revealed three polymorphic sites between Kr-0 and Col-0 . Two of these are SNPs: one in an intron , and the other a synonymous SNP in an exon ( Table S2 ) . The other polymorphism is a 1-bp insertion in the second exon in Kr-0 causing replacement of Leu53 with a Thr and introducing a premature stop codon ( Figure 2F ) that truncates the protein by 116 amino acids . This truncation likely results in loss-of-function making this 1-bp insertion in At2g21045 responsible for high leaf arsenic in Kr-0 . Loss-of-function alleles of HAC1 were unable to complement the high leaf arsenic of Kr-0 in F1 hybrids ( Figure 2G ) , whereas the Col-0 At2g21045 genomic fragment introduced transgenically into Kr-0 was able to complement ( Figure 2H ) , confirming that At2g21045 is the causal gene for HAC1 . At2g21045 was named HAC1 . An analysis of HAC1 in 220 re-sequenced A . thaliana accessions , and the Sanger sequencing of HAC1 in five accessions with high leaf arsenic ( App1-16 , Benk-1 , Bur-0 , Ann-1 , Hn-0 ) identified from our screen of 349 accessions , did not identify the 1 bp insertion observed in Kr-0 in any other accessions . This suggests that there is still significant functional allelic diversity at HAC1 contributing to variation in leaf arsenic that needs to be described . We observed that the predicted amino acid sequence encoded by HAC1 contains a Rhodanase-like domain ( IPR001763 ) , a domain known to exist in arsenate reductase enzymes [61] , in yeast [21] , and in plants [22] , [24] , [25] . This suggested the possibility that HAC1 functions as an arsenate reductase , reducing arsenate ( AsV ) to arsenite ( AsIII ) . To test this hypothesis , we evaluated the oxidation state of arsenic in plants lacking a functional HAC1 gene . Consistent with HAC1 encoding an arsenate reductase , both hac1-1 and hac1-2 loss-of-function alleles have increased accumulation of arsenate in shoots and roots compared to wild-type Col-0 ( Figure 3A , 3B ) . Kr-0 also has increased arsenate accumulation in both shoots and roots , consistent with HAC1Kr-0 being a loss-of-function allele . Moreover , in roots of hac1-1 , hac1-2 , and Kr-0 we observe a reduction in the percentage of the total accumulated arsenic present as arsenite . In roots of Col-0 wild-type plants 98% of the total accumulated arsenic is present as arsenite , whereas in roots of plants with a loss-of-function allele of HAC1 ( hac1-1 , hac1-2 , and Kr-0 ) arsenite accounts for between 79%–83% of total root arsenic . This reduced capacity to convert arsenate to arsenite is also associated with a significant increase in arsenic , primarily as arsenite , in the shoots ( Figure 3C ) . We propose this increase in arsenic in shoots is driven by the enhanced accumulation of arsenate in the roots ( Figure 3B ) , which is then readily translocated to the shoot . Once in the shoot this arsenate is reduced to arsenite via HAC1-independent mechanisms . As would be expected for an arsenate reductase , loss-of-function of HAC1 had no effect on the accumulation or oxidation state of arsenite when plants were exposed to arsenite in the growth medium ( Figure S1A , S1B ) . To further test the hypothesis that HAC1 encodes an arsenate reductase , we heterologous expressed HAC1Col-0 in a strain of Escherichia coli lacking its endogenous ArsC arsenate reductase . E . coli lacking ArsC is known to have enhanced sensitivity to arsenate toxicity because , without ArsC , arsenic is unable to be extruded from cells as arsenite , causing an accumulation of cellular arsenic that leads to enhanced arsenate sensitivity . We observed that heterologous expression of HAC1 in the ΔarsC mutant lacking a functional copy of arsC suppressed this enhanced sensitivity to arsenate ( Figure 4A ) , in a similar manner to other known plant and yeast arsenate reductases [23] , [24] , [62] . This suppression of arsenate sensitivity by HAC1Col-0 was also associated with a recovery of the ability of the ΔarsC E . coli mutant to efflux arsenite back into the growth medium ( Figure 4B ) . As expected , arsenate reductase activity is almost undetectable in a cell free extract of the E . coli ΔarsC mutant lacking an arsenate reductase ( Figure 4C ) . However , heterologous expression of the A . thaliana HAC1Col-0 gene in the E . coli ΔarsC loss-of-function mutant confers the ability to reduce arsenate to arsenite ( Figure 4C ) . In this assay , arsenate reductase activity is monitored as the loss of the primary electron donor NADPH through its conversion into NADP+ during the coupled reduction of arsenate to arsenite . Plants are known to contain enzymes with arsenate reductase activity when tested in heterologous systems [22] , [24] , [25] , and that shows sequence homology to the yeast arsenate reductase ACR2 ( Figure 5A ) . These previously characterised plant enzymes contain the conserved HCX5R active site [63] found in the yeast ACR2 ( Figure 5B ) , but these plant enzymes appear not to impact arsenic metabolism [26] . In contrast , HAC1 in A . thaliana and its homologs in rice do not contain the yeast ACR2 canonical arsenate reductase active site ( Figure 5B ) . HAC1 is well conserved in plants ( Figure S2 ) , with several domains of the protein having very high levels of conservation . However , further work is required to understand the functional significance of these highly conserved domains . We used reciprocal grafting to determine that a functional HAC1 is required in roots to maintain a low arsenic concentration in shoots . Only plants with wild-type Col-0 roots containing a functional HAC1 allele showed Col-0 like shoot arsenic concentrations , even when Col-0 roots are grafted onto Kr-0 shoots that contain a non-functional HAC1 allele ( Figure 6A ) . This root function of HAC1 is consistent with HAC1 being primarily expressed in roots ( Figure 6B ) . Furthermore , transformation of Col-0 wild type with a HAC1Col-0-Green Fluorescent Protein ( HAC1Col-0-GFP ) construct expressed from the HAC1 promoter allowed the detection of the HAC1-GFP fusion protein , which we observed to be localised to root hairs and epidermal cells at the surface of the root and to the pericycle within the stele ( Figure 6C , 6E and Figure S3 ) . Exposure to arsenate in agar solidified growth medium for 3 days causes a significant , dose dependent , increase in the steady state levels of HAC1 mRNA in roots of Col-0 wild type ( Figure 7A ) . This increase in HAC1 mRNA levels reaches a maximum at 100 µM arsenate in the growth media ( Figure 7A ) . In contrast , exposure to arsenite under the same conditions significantly reduces HAC1 mRNA levels ( Figure 7B ) at all concentrations tested . To further understand the physiological role of HAC1 catalysed arsenate reduction in roots , we analysed both arsenite efflux and arsenate uptake by roots in hac1 loss-of-function mutants and Kr-0 ( Figure 8 ) . To avoid artefacts caused by arsenic toxicity , these experiments were performed at 5 µM arsenate in the hydroponic nutrient solution , a subtoxic concentration of arsenate . We observed that arsenite efflux from roots , measured as the accumulation of arsenite in the nutrient media after exposure of the plants to arsenate , was dramatically reduced in both the HAC1 loss-of-function mutants and Kr-0 compared to Col-0 wild-type plants ( Figure 8A ) . Col-0 wild-type plants extruded 2 µmoles arsenite g−1 fresh weight of roots over a 48 hr period after exposure to arsenate . In contrast , in both hac1-1 and hac1-2 this efflux of arsenite was reduced 10-fold and was also reduced to a similar level in Kr-0 , which also contains a loss-of-function allele of HAC1 . On the other hand , arsenate uptake , measured as depletion of arsenate from the nutrient solution , was no different in hac1-1 , hac1-2 , or Kr-0 compared to wild-type Col-0 at both time points tested ( Figure 8B ) . Such evidence establishes that HAC1 is essential for arsenite efflux from roots , but is not necessary for arsenate uptake . To understand the role of HAC1-facilitated arsenite efflux in limiting arsenic translocation to the shoot , we investigated the distribution of arsenic in roots in the hac1-1 and hac1-2 loss-of-function mutants . Using Synchrotron μX-ray Fluorescence ( μ-XRF ) we were able to image arsenic accumulation at µm resolution in high pressure frozen , freeze-substituted sections of roots . Such imaging revealed that in the absence of a functional HAC1 gene , and with the loss of arsenite efflux , arsenic appears to accumulate mainly in the root stele to concentrations 120%–180% higher than in the Col-0 wild type ( Figure 9 ) . The arsenic concentration in the scanned root sections of wild-type , hac1-1 and hac1-2 ( Figure 9B ) reached a maximum of 411 , 917 , and 1 , 154 mg kg−1 , respectively . We demonstrate that HAC1 plays a critical role in controlling the amount of arsenic accumulated in roots and shoots in A . thaliana . To evaluate if such a role is important in arsenic resistance , we exposed both hac1-1 and hac1-2 loss-of-function mutants to increasing concentrations of arsenate or arsenite in the agar solidified growth medium . Exposure to either arsenate ( Figure 10 ) or arsenite ( Figure S4 ) caused a significant reduction in both root growth and overall plant fresh weight ( Figure 10 and Figure S4 ) for all genotypes . However , hac1-1 and hac1-2 were both significantly more sensitive to arsenate compared to the Col-0 wild type at relatively high concentrations of arsenate when growth was measured as root length or overall plant fresh weight ( Figure 10 ) . However , we observed no significant difference in sensitivity to arsenite between hac1-1 , hac1-2 , and Col-0 wild type ( Figure S4 ) . This observation supports our conclusion that HAC1 acts as an arsenate reductase in roots that controls arsenic accumulation in both roots and shoots when plants are exposed to arsenate . Previously , ACR2 ( also known as CDC25 ) from A . thaliana has been shown to encode an arsenate reductase when assayed in vitro [25] . We were , therefore , interested to know if HAC1 and ACR2 interact epistatically . To test this , we generated the double mutant acr2 hac1 homozygous for loss-of-function alleles of both ACR2 and HAC1 . The acr2-2 hac1-2 double mutant , along with both the parental single mutants and wild-type background , was exposed to arsenate in agar solidified growth medium or in hydroponic nutrient solution and the phenotypes associated with arsenic metabolism ( arsenate uptake , arsenate reduction , arsenite efflux , arsenic accumulation , and arsenate resistance ) evaluated . After exposure to 5 µM arsenate for 24 hr in the hydroponic medium , we observed in the hac1-2 single mutant a significant increase in the accumulation of arsenate in roots ( Figure 11A ) and a significant decrease in arsenite efflux ( Figure 11D ) , leading to a significant increase in total arsenic accumulation in shoots ( Figure 11B ) with no effect on arsenate uptake ( Figure 11C ) . This is what we had previously observed for the single hac1 mutants ( Figure 3 ) . Furthermore , loss-of-function of ACR2 had no effect on the accumulation of arsenate in roots , efflux of arsenite or uptake of arsenate , or the total arsenic accumulation in shoots , as previously published [26] . Combining both acr2-2 and hac1-2 loss-of-function alleles in the acr2-2 hac1-2 double mutant did not significantly alter either arsenate accumulation in roots or arsenite efflux compared to the hac1-2 single mutant ( Figure 11A , 11D ) . We did observe a significant increase in shoot arsenite accumulation in the acr2-2 hac1-2 double ( Figure 11B ) . However , this was not observed in the alternative acr2-2 hac1-1 double mutant ( Figure S5B ) , and we therefore conclude that ACR2 and HAC1 do not interact epistatically under the conditions in which we tested the phenotypes . Furthermore , at all arsenate concentrations tested no significant difference was observed in the growth ( root length or overall fresh weight ) of Col-0 wild type and acr2-2 ( Figure 12 ) . This is similar to results previously published [26] and suggests that ACR2 is not necessary for arsenate reduction or resistance under the conditions used in our experiments . As expected , loss-of-function of HAC1 significantly reduced arsenate resistance compared to Col-0 wild type , measured as either root length or overall fresh weight ( Figure 12 ) at concentrations of arsenate at and above 50 µM . However , combining both acr2-2 and hac1-2 loss-of-function alleles in the acr2-2 hac1-2 double mutant did not significantly alter the arsenate resistance compared to the hac1-2 single mutant ( Figure 12 ) . From these results , we conclude that under the conditions we tested HAC1 and ACR2 do not interact epistatically to affect arsenic metabolism or resistance . Furthermore , we reconfirm that ACR2 plays no observable role in arsenic metabolism or resistance in vivo [26] .
Here , we have used genome-wide association mapping of loci associated with leaf arsenic accumulation to identify HAC1 , a gene encoding a protein that plays a critical role in reducing arsenate to arsenite in roots to promote arsenite efflux as part of a plant's arsenic resistance mechanism ( Figure 13 ) . While this paper was in review , Sánchez-Bermejo and colleagues [64] also identified this gene , which they named ATQ1 , as an arsenate reductase playing a critical role in arsenate resistance . However , in contrast to our study , these authors used QTL mapping of loci linked to variation in arsenate resistance in a biparental RIL population to identify ATQ1 . We go beyond the findings of these authors by revealing the functional role of ATQ1 in arsenate resistance and investigating the role of HAC1/ATQ1 in arsenic accumulation . We find that natural variation at the HAC1 locus accounts for a significant proportion of the species-wide diversity in leaf arsenic accumulation in A . thaliana when plants are grown in soil containing environmentally relevant trace concentrations of arsenic . Furthermore , we identify the A . thaliana accession Kr-0 , collected from the Botanic Garden in Krefeld , Germany , as having a rare natural loss-of-function allele of HAC1 that leads to extreme foliar accumulation of arsenic in this accession . We show that the HAC1 protein accumulates in the root epidermis and root hairs , where it is ideally localised to play a critical role in the chemical reduction of arsenate to arsenite , a role necessary to allow arsenic , as arsenite , to be extruded from roots . Such efflux of arsenite is vital in order to control arsenic accumulation , which , left unchecked , can cause arsenic hyperaccumulation and toxicity . Arsenite efflux represents a large proportion of the arsenic taken up as arsenate by roots ( Figure 8 and Figure 11 ) [26] . In the absence of functional HAC1 protein this efflux is abolished and arsenic over-accumulates in roots and shoots , leading to arsenic toxicity . In E . coli arsenate resistance is achieved in a similar manner to plants , by the reduction of arsenate to arsenite and efflux of arsenite from the cell [65] . We observed that the HAC1 gene from A . thaliana is able to restore arsenate reduction capacity , arsenite efflux , and arsenate resistance to E . coli lacking their endogenous arsenate reductase ( ArsC ) , in a similar manner to that recently shown by Sánchez-Bermejo and colleagues [64] . Furthermore , Sánchez-Bermejo and colleagues verified that the purified HAC1/ATQ1 recombinant protein has arsenate reductase activity [64] . These observations , taken together with the fact that arsenate accumulates in roots of plants lacking HAC1 , strongly suggests that HAC1 encodes an arsenate reductase enzyme . In the absence of a functional HAC1 protein , A . thaliana still maintains the ability to reduce substantial amounts of arsenate to arsenite , suggesting that there are other arsenate reduction mechanisms in the plant . This redundancy has also been observed in E . coli [66] and S . cerevisiae [26] . One possibility is that arsenate could be reduced non-enzymatically by glutathione , though conversion rates might be slow . However , enzymes forming phosphorylated products can also promote arsenate reduction by incorrectly using arsenate in place of phosphate to generate arsenylated products in which the arsenate is more easily reduced by thiols such as glutathione [67] . Furthermore , both glutaredoxin and triosephosphate isomerise [68] , [69] have also been suggested to promote arsenate reduction by as-yet-unknown mechanisms . More importantly , what is non-redundant for ArsC in E . coli , ACR2 in S . cerevisiae , and HAC1 in A . thaliana is their function as arsenate reductases enabling arsenite efflux and resistance to arsenate . Such observations suggest that these arsenate reductases are necessary for the reduction of arsenate to arsenite to create a specific pool of arsenite for efflux . It has been proposed [66] that this specificity may come about by the direct physical interaction of the arsenate reductase with the arsenite effluxer , thereby channelling arsenic for efflux . In support of this an actinobacterial enzyme containing an aquaglyceroporin-derived arsenite channel with a C-terminal arsenate reductase has been identified that provides single gene arsenate resistance [70] . The coupling of arsenate reduction and arsenite efflux could also occur by both proteins being localised to the same specific cellular location where arsenite produced by the reductase could be efficiently channelled to the efflux protein . In mutants lacking a functional HAC1 non-specific arsenate reduction may occur at locations that do not contain the arsenite efflux protein , leading to arsenate reduction without efflux . Such suggested coupling of the HAC1 , ArsC , and ACR2 arsenate reductases to their associated arsenite effluxer is quite different from the previously characterised A . thaliana arsenate reductase ACR2 . This enzyme has been shown to have arsenate reductase activity as a purified enzyme in vitro [25] . However , loss-of-function of ACR2 has no effect on arsenate reduction , arsenite efflux , arsenic accumulation or resistance to arsenate in vivo , as previous published [26] and repeated here . Furthermore , our genetic analysis using the acr2 hac1 double mutant reveals that ACR2 also does not interact epistatically with HAC1; supporting our conclusion that ACR2 plays no role in arsenic metabolism in A . thaliana , even though in vitro ACR2 has arsenate reductase activity [25] . Our work goes beyond the recently published work of Sánchez-Bermejo and colleagues [64] in exploring experimentally the genetic interaction between HAC1/ATQ1 and ACR2 . In the hac1 A . thaliana mutant that lacks the arsenate reductase activity required for arsenite efflux , we observe an over-accumulation of arsenic within the central stele of the root . This suggests that in the absence of sufficient arsenite efflux , due to a lack of appropriate arsenate reduction capacity , arsenate is transported radially across the root and accumulates within the stele ( Figure 13 ) . Once in the stele , this excess arsenate would be expected to load into the xylem and be translocated to the shoots , where it would be reduced to arsenite by HAC1 independent mechanisms , leading to arsenic hyperaccumulation , as observed in the A . thaliana hac1 mutant . Interestingly , we also observe the HAC1 protein to be localised to the pericycle close to the xylem within the stele . HAC1 in this location could provide the arsenate reduction capacity needed to maintain low concentrations of arsenate in the stele ( Figure 13 ) . Our work goes beyond the recently published work of Sánchez-Bermejo and colleagues [64] in developing such mechanistic insights . Since arsenate is known to act as a phosphate analogue , and can therefore be potentially loaded into the xylem via the phosphate transport system , limiting arsenate concentrations in the stele would be an effective mechanism to minimise arsenic translocation to the shoot . PHO1 is the main phosphate effluxer involved in xylem loading of phosphate [71] , [72] in A . thaliana . Loss-of-function of PHO1 does not reduce arsenic accumulation in shoots [73] , and this supports the notion that in wild-type plants , with a functional HAC1 protein , arsenate concentrations in the stele are low enough that arsenate translocation does not contribute significantly to shoot arsenic accumulation [73] . In plants with a non-functional HAC1 , the expected elevated translocation of excess arsenate does not appear to compete with phosphate transport , since leaf phosphate concentrations are equal in the wild type and hac1 mutant ( Figure S6 ) . However , here plants were grown in soil containing only 0 . 1 µmoles g−1 dry weight arsenate and plants were fertilised regularly with nutrient solution containing 0 . 25 mM phosphate , making it unlikely that arsenate competition with phosphate for xylem loading could be detected . Competition may , however , be detectable if plants are exposed to high concentrations of arsenate , though such exposure would be toxic to the plant . Loss-of-function of HAC1 causes increased shoot accumulation of arsenic when plants are growing in soils containing only 7 . 5 mg/kg arsenic , a concentration well below many countries' clean-up guidelines for soil arsenic , which range from 0 . 039–40 mg/kg in the US ( depending on the state ) to 150 mg/kg in Japan ( 15 mg/kg limit applies to rice fields ) [74] . This suggests that HAC1 functions constitutively to allow plants to maintain control over arsenic accumulation even when they are growing in soils that contain low arsenic concentration . However , HAC1 expression is also induced by exposure to arsenate . This suggests that HAC1 is involved in both constitutive and plastic responses of A . thaliana to arsenate in the environment , allowing plants to maintain effective arsenite efflux from roots across a range of arsenic concentrations in the soil . The discovery of HAC1 now provides an explanation of why T-DNA insertion alleles of ACR2 have been reported to not affect arsenic homeostasis in A . thaliana [26] , whereas RNA interference has been shown to have a strong effect [23] . Since ACR2 and HAC1 share sequence identity ( Figure S7 ) within the region used to knock down expression of ACR2 by RNA interference [23] , this sequence may also have suppressed HAC1 expression . Such suppression of HAC1 expression would then explain the enhanced arsenate sensitivity and arsenic hyperaccumulation observed by Dhankher and colleagues [23]; phenotypes that are the same as those we observed for the hac1 mutants in A . thaliana . Close to 18% of the total variation in leaf arsenic observed in the 349 A . thaliana accessions tested is explained by SNPs marking the linkage block that contains HAC1 , and plants with the major allele of this locus contain , on average , 37% higher leaf arsenic concentration . The Kr-0 accession contains the major allele for this locus , and the elevated leaf arsenic in this accession is explained by the presence of a loss-of-function allele of HAC1 . However , the nucleotide polymorphism in HAC1Kr-0 that generates a non-functional HAC1 protein is not present in any other accession screened to date , establishing it as a rare allele . However , given that almost 75% of the accessions used in our GWA analysis contain the high leaf arsenic allele of the HAC1 QTL , a significant amount of allelic diversity within the linkage block containing the HAC1 gene remains to be characterised . The Kr-0 HAC1 allele may therefore represent an extreme allele from within a larger set of HAC1 alleles with less extreme variation in function . While this paper was in review , Sánchez-Bermejo and colleagues [64] reported on the existence of a second weak allele of HAC1 ( which they called ATQ1 ) from the Kas-1 accession . Unlike HAC1Kr-0 , which contains a premature stop codon , the Kas-1 allele contains multiple non-additive intra-allelic polymorphisms that lead to its reduced function . It is also interesting to note that since the majority of the A . thaliana accessions tested have the HAC1 allele associated with higher leaf arsenic , they are likely to have weak alleles of HAC1 . However , the selective benefit , if any , of having this weak allele of HAC1 remains an open question . The close coupling of arsenate reduction by HAC1 with arsenite efflux from roots suggests that both HAC1 and the arsenite effluxer [14] are closely associated through perhaps co-expression in the same cells and possibly via direct protein–protein interactions . Critically , it is this arsenite efflux process in roots that allows plants to maintain low arsenic concentrations in their shoots when they are exposed to environmentally relevant arsenate concentrations in the soil ( Figure 13 ) . The discovery of HAC1 and its role in this process opens up new possibilities for the development of crop plants with reduced arsenic concentrations in their edible parts , potentially providing real benefits to human health by limiting arsenic intake in the diet . The existence of natural genetic variation at HAC1 in A . thaliana holds out the promise that variation in HAC1 function may also exist in crops , providing the genetic material to develop low-arsenic-accumulating varieties . However , given the high frequency of the weak allele of HAC1 across the A . thaliana species , it will be important to understand if there is any negative trade-off to having highly efficient arsenate reduction capacity .
The set of A . thaliana accessions used for this study contained 349 accessions selected from 5 , 810 worldwide accessions as previously described [44] , [46] . T-DNA insertion mutants ( GABI_868F11 , SM_3_38332 for hac1-1 and hac1-2 , and GABI_Kat772G06 for acr2-2 ) were obtained from the Nottingham Arabidopsis Stock Centre ( NASC ) . The plants used for analysis of leaf arsenic by ICP-MS were grown for 5 wk in a climate-controlled room with a photoperiod of 10 hr light ( 90 µmol m−2 s−1 ) 14 hr dark , humidity of 60% and temperature ranging from 19 to 22°C . The soil preparation , seed stratification , sowing , and plant cultivation followed protocols described previously [44] , [46] . Plants used for analysis of the expression of HAC1 , including quantitative real-time RT-PCR and for cell type and tissue expression pattern ( pHAC1::HAC1-GFP transgenic lines ) , were grown in axenic conditions . In brief , seeds were surface sterilized with 50% bleach and 0 . 05%SDS for 15 min followed by being washed eight times with sterilized , deionized water and sown on 1/2 strength Murashige and Skoog ( Sigma-Aldrich , St . Louis , US ) media solidified with agar containing 1% sucrose in Petri dishes . Seeds on plates were stratified at 4°C for 3 days . Plates with seeds were then maintained at 16 hr light ( 90–120 µmol m−2 s−1 ) and 8 hr dark at 22°C . Hydroponic experiments were carried out as previously described [26] . Three-week-old plants were exposed to 5 µM arsenate for 24–48 hr . Arsenic species in the nutrient solution and in the roots and shoots were determined by HPLC-ICP-MS . The decrease of arsenate and the production of arsenite in the medium were used to calculate the arsenate uptake and arsenite efflux , respectively [26] . The concentration of total leaf arsenic was measured as 75As using inductively coupled plasma mass spectrometry ( ICP-MS ) as described previously [75] . Briefly , one to two adult rosette leaves were harvested from 5-week-old A . thaliana plants . The leaves were cleaned by rinsing with ultrapure water ( 18 . 2 MΩcm Milli-Q , Merck Millipore ) and placed into Pyrex digestion tubes . Samples were dried in an oven at 88°C for 20 hours . After cooling , seven reference samples from each planted block were weighed . The samples , together with blank controls , were digested with 0 . 90 ml concentrated nitric acid ( Baker Instra-Analyzed; Avantor Performance Materials ) and diluted to 10 . 0 ml with ultrapure water ( 18 . 2 MΩcm ) . The internal standard Indium ( In ) was added to the acid prior to digestion for monitoring technical errors and plasma stability in the ICP-MS instrument . After samples and controls were prepared , elemental analysis was performed with an ICP-MS ( Elan DRC II or NexION 300D; PerkinElmer ) coupled to Apex desolvation system and SC-4 DX autosampler ( Elemental Scientific Inc . , Omaha , NE , US ) , monitoring these elements: Li , B , Na , Mg , P , S , K , Ca , Mn , Fe , Co , Ni , Cu , Zn , As , Se , Rb , Sr , Mo , and Cd . All samples were normalized to calculate weights , as determined with a heuristic algorithm using the best-measured elements , the weights of the seven weighed samples , and the solution concentrations [75] , detailed at www . ionomicshub . org . For GWA analysis , data were normalised using common genotypes across experimental blocks as previously described [46] , [76] , and these normalised data have been deposited on the iHUB ( previously known as PiiMS [77] ) for viewing and download through www . ionomicshub . org . Of the set of 349 A . thaliana accessions analyzed for leaf arsenic , a subset of 337 accessions were genotyped for 213 , 497 SNPs using the custom-designed SNP-tilling array Atsnptile 1 [44] , [46] , [78] . The GWA analysis was performed using a linear mixed model to correct confounding by population structure [79] implemented in the program EMMA ( Efficient Mixed-Model Association ) described previously [28] . XAM was done as described previously [44] , [59] . In brief , phenotyped F2 individuals were sorted by leaf arsenic concentration and approximately one quarter at each end of the F2 population distribution were pooled separately . Genomic DNA was extracted from the two pools and labelled separately using the BioPrime DNA labelling system ( Invitrogen ) . The labelled DNA was hybridized to the Affymetrix SNP-tilling array Atsnptile 1 . The CEL files containing raw data of signal intensity for all probes were read and spatially corrected using previously described R scripts [80] with the R program and the Bioconductor Affymetrix package . The original CEL files used in this study can be found in the Gene Expression Omnibus ( GEO ) under accession GSE62299 . There are antisense and sense probes for each of the previously characterized polymorphic diallelic SNPs used here as genetic markers . The allele frequency difference between the two pools for each of these SNP markers was scored based on the signal intensity difference of the probes . The whole process was carried out using R scripts described previously [59] . The mapping interval by XAM was further narrowed down by PCR-based genotyping . Firstly , the 315 individuals of the F2 population used for XAM were genotyped individually at six cleaved-amplified polymorphic sequence ( CAPS ) markers as indicated in Figure 1C . Recombinants between marker CS85HKA and CS95HKB were selected for further analysis . The F2 recombinants with leaf arsenic concentration similar to Kr-0 were directly used for determination of the candidate region based on recombination between phenotype and genotypes at each marker , while the recombinants similar to Col-0 were selfed and the location of the candidate gene was determined based on recombination between phenotype and genotypes at each marker in the F3 progeny . After the rough mapping , three more polymorphic markers were developed within the rough mapping region between CS8901K and CS9249K , and an enlarged F2 population with 1 , 321 F2 individuals was used for fine mapping of the candidate gene . The candidate gene was further narrowed down using the same strategy as used for rough mapping . The primers and restriction enzymes for the CAPS markers are listed in Table S3 . The HAC1 genomic region of Kr-0 was sequenced using overlapping PCR as described previously [44] . Briefly , four pairs of primers for the PCR reactions were designed using Overlapping Primersets ( http://pcrsuite . cse . ucsc . edu/Overlapping_Primers . html ) ( Table S3 ) , and four overlapping fragments were amplified using these four pairs of primers with Kr-0 genomic DNA as the template . Each fragment was sequenced using its amplification primers in two directions . The sequenced reads were assembled using SeqMan Lasergene software ( DNASTAR; http://www . dnastar . com ) , with the Col-0 sequence as the reference , and polymorphisms were identified by comparing the reference sequence and the Kr-0 sequence . To construct the complementation vector of HAC1 , a genomic DNA fragment including 1 . 49 kb promoter region , gene body and 0 . 34 kb 3′ downstream sequence was amplified by PCR from Col-0 using KOD hot start DNA polymerase ( TOYOBO Bio-Technology , CO . , LTD , Japan ) and primer HAC-CU and HAC-CL ( Table S3 ) . The fragment was cloned into the pCR-XL-TOPO vector ( Invitrogen Life Technologies ) for sequencing and subsequently introduced into the binary vector pHB [81] using restriction enzymes EcoR I and Pst I to replace the 2×35S promoter . To construct the expression vector for expressing the fusion protein of HAC1-GFP driven by HAC1 promoter , the HAC1 genomic fragment including 1 . 49 kb promoter region and gene body with the stop codon replaced with TTA was PCR amplified from Col-0 using primer HAC-CU and HAC-GFP-Linker1 ( Table S3 ) . The GFP coding fragment was amplified from the pMDC vector using primer pair HAC-GFP-Linker2 and GFP-RP ( Table S3 ) . Thereafter , the fusion fragment of pHAC1:HAC1-GFP was amplified using the primer pair HAC-CU and GFP-RP with the above two PCR fragment products as template . The pHAC1:HAC1-GFP fragment was cloned into pCR-XL-TOPO vector for sequencing , and cloned into the pHB plant expression vector [81] using the EcoR I and Pst I restriction sites in pHB . The expression vectors were transformed into Agrobacterium tumefaciens strain GV3101 and introduced into Kr-0 and Col-0 using the floral dip method [82] . The transgenic lines were screened on half-strength Murashige and Skoog ( Sigma-Aldrich , St . Louis , US ) agar plates containing 50 µg/ml Hygromycin and 1% sucrose . For prokaryotic expression of His-tagged HAC1 , the full-length coding sequence of HAC1 was amplified using the primers HAC-PEF and HAC-PER ( Table S3 ) and cDNA products reverse-transcribed from Col-0 root RNA as the template . The fragment was cloned into T-easy vector ( Promega ) . As the forward primer HAC-PEF introduced a Nde I restriction site , the insertion direction of the fragment in each clone was tested with the restriction enzyme Nde I . The clones with a Nde I site in the forward primer and a Nde I site in the multiple cloning site of the vector flanking the fragment were sequenced . The correct fragment was then cloned in-frame into the Nde I site of the prokaryotic expression vector pColD-TF ( Takara , Japan ) , and verified by sequencing . The vector was transformed into E . coli ΔarsC mutant WC3110 ( a strain lacking arsenate reductase activity ) [62] , [83] and its wild-type W3110 for complementation and assaying arsenate reductase activity . Expression of the His-tagged HAC1 was induced with 1 mM IPTG at 16°C for at least 16 hr . Reciprocal grafting was performed as previously described [44] . After graft unions established , grafted plants were examined under the stereoscopic microscope before transfer to potting mix soil to identify any adventitious root formation from the graft unions or above . Healthy grafted plants without adventitious roots were transferred to potting mix soil and grown in a controlled environment described above . After 4 wk , leaf samples were harvested for arsenic analysis . After harvesting , plants were examined again , and those with adventitious roots or without a clear graft union were removed from the subsequent analysis of the arsenic data . The ΔarsC mutant WC3110 and its wild-type W3110 with pColD-TF empty vector or pColD-TF-HAC1 were cultured at 37°C overnight . All cultured strains were diluted to OD600 nm = 0 . 5 , and 50 µl inoculated into 5 ml of LB liquid media containing 1 mM IPTG and different concentrations of arsenate , as indicated in Figure 2 . Cells were cultured at 16°C for 72 hr or at times indicated . The cell density was measured at OD600 nm using a spectrophotometer . The E . coli bacteria strains for expressing His-tagged HAC1 were lysed in arsenate reductase assay buffer ( 50 mM MES , 50 mM MOPS , pH 6 . 5 , 300 mM NaCl , 0 . 1 mg/mL bovine serum albumin and 1% proteinase inhibitor cocktail [Sigma P9599] ) using a French Press with a low temperature ultra-high pressure continuous flow cell disrupter ( JNBIO Co . , Ltd , China ) . The cell lysate was centrifuged at 16 , 000 g for 10 min at 4°C to remove cell debris and unbroken cells . Total protein concentration was measured using Coomassie Plus protein assay reagent ( Pierce 23236 ) . Arsenate reductase activity of the cell free extracts was measured using the previously established coupled assay [62] . Arsenite effluxed to the growth medium was measured by HPLC-ICP-MS . Plant samples were ground in liquid nitrogen to a fine powder in a mortar and pestle . The finely ground material ( approximately 0 . 1 g ) was extracted with 10 ml phosphate buffer solution ( 2 mM NaH2PO4 and 0 . 2 mM Na2-EDTA , pH 5 . 5 ) for 1 hr under sonication in a 4°C cold room [84] . The extract was filtered firstly through No . 42 Whatman filter paper and then through a 0 . 2 µm membrane filter . Arsenic speciation was determined using HPLC-ICP-MS ( PerkinElmer NexION 300× , Waltham , MA , US ) . Arsenic species were separated using an anion-exchange column ( Hamilton PRP X-100 , fitted with a guard column; Reno , NV , US ) with a mobile phase of 6 . 0 mM NH4H2PO4 , 6 . 0 mM NH4NO3 , and 0 . 2 mM Na2-EDTA ( pH 6 . 2 ) , run isocratically at 0 . 7 ml min−1 . The solution from the separation column was mixed continuously with an internal standard solution ( Indium ) before being introduced into the ICP-MS . The instrument was set up in the kinetic energy discrimination mode with helium as the collision gas to reduce polyatomic interferences . Signals at m/z 75As and 115In were collected with a dwell time of 300 ms; the In counts were used to normalise the As counts . Arsenic species in the samples were quantified by external calibration curves using peak areas . The pHAC1::HAC1-GFP transgenic A . thaliana lines grown in axenic condition for 1 wk were used for microscopic observation . GFP fluorescence of seedlings was observed using a stereo fluorescence microscope ( M165FC , Leica ) and a confocal laser microscope ( FV-1000 , OLYMPUS ) . Total RNA extraction and cDNA synthesis were performed as described previously [44] . Quantitative real-time RT-PCR was done using a Real-Time PCR System ABI StepOnePlus ( Life Technologies , US ) using SYBR Green PCR Master Mix ( Life Technologies , US ) with the first strand cDNA as a template . Primers for quantitative RT-PCR ( Table S3 ) were designed using Primer Express Software Version 3 . 0 ( Life Technologies , US ) with one primer of a pair covering an exon–exon junction . Expression data analysis was performed as previously described [44] . Phylogenetic analyses were conducted using MEGA version6 [85] . Protein sequences were aligned using MAFFT7 . 1 [86] and the tree constructed using the parsimony method . Bootstraps were carried out with 1 , 000 replications . The GeneBank accession numbers for the protein sequences or the nucleotide sequences from which protein sequence were derived are: AY860059 ( OsACR2-1 ) , AY860058 ( OsACR2-2 ) , DQ310370 ( PvACR2 ) , BT003658 . 1 ( AtACR2 ) , BT008306 . 1 ( AtHAC1 ) , BAD07813 . 1 ( OsHAC1-1 ) , NP_001052310 . 1 ( OsHAC1-2 ) , NP_014928 . 1 ( ScHAC1 ) , CAB83305 . 1 ( AtACR2-2 ) , YP_005275964 . 1 ( AsrC ) . Three-week-old seedlings of Col-0 , hac1-1 , and hac1-2 were exposed to 10 µM arsenate for 10 days in a hydroponic culture . Segments of roots at approximately 2 cm from the root tip were cut and placed into a planchette coated with hexadecane . The samples were frozen at −196°C with a pressure of 210 MPa for 30 s using a Leica HPM100 high pressure freezer [87] . The frozen samples were freeze substituted , embedded in resin , and sectioned into 7 µm thickness as previously described [87] . Synchrotron μXRF was undertaken at the Diamond Light Source on the I18 microfocus beamline . The incident X-ray energy was set to 12 . 4 keV using a Si ( 111 ) monochromator . The X-ray fluorescence spectra were collected using a Si drift detector . The beam size and step size were both 2 µm . Quantification of the concentrations of arsenic and other elements of interest in the samples were carried out using an external calibration with XRF reference materials . | Arsenic is a human carcinogen that accumulates from soil into many different food crops , where it presents a significantly increased cancer risk when foods derived from these crops are consumed . Plants naturally control the amount of arsenic they accumulate by first chemically converting arsenate into arsenite , which is then extruded from the roots back into the soil . Because arsenate is a chemical analogue of phosphate , conversion of arsenate in the root to arsenite may also prevent arsenic being efficiently transported to the shoots via the phosphate transport system . The chemical reduction of arsenate to generate arsenite is therefore clearly a key component of a plant's detoxification strategy . Here , we use genetic methods to identify the enzyme responsible for this crucial reaction—HAC1 . We show that HAC1 is responsible for arsenate reductase activity in both the outer layer of the root ( epidermis ) and the inner layer adjacent to the xylem ( pericycle ) . In its absence , the roots return less arsenic to the soil and the shoots accumulate up to 300 times more arsenic . This knowledge creates new opportunities to limit arsenic accumulation in food crops , thereby helping to reduce the cancer risk from this food-chain contaminant . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] | [
"biotechnology",
"biochemistry",
"plant",
"biochemistry",
"plant",
"science",
"plant",
"genomics",
"genetics",
"biology",
"and",
"life",
"sciences",
"plant",
"genetics",
"plant",
"biotechnology"
] | 2014 | Genome-wide Association Mapping Identifies a New Arsenate Reductase Enzyme Critical for Limiting Arsenic Accumulation in Plants |
Results from Genome-Wide Association Studies ( GWAS ) have shown that complex diseases are often affected by many genetic variants with small or moderate effects . Identifications of these risk variants remain a very challenging problem . There is a need to develop more powerful statistical methods to leverage available information to improve upon traditional approaches that focus on a single GWAS dataset without incorporating additional data . In this paper , we propose a novel statistical approach , GPA ( Genetic analysis incorporating Pleiotropy and Annotation ) , to increase statistical power to identify risk variants through joint analysis of multiple GWAS data sets and annotation information because: ( 1 ) accumulating evidence suggests that different complex diseases share common risk bases , i . e . , pleiotropy; and ( 2 ) functionally annotated variants have been consistently demonstrated to be enriched among GWAS hits . GPA can integrate multiple GWAS datasets and functional annotations to seek association signals , and it can also perform hypothesis testing to test the presence of pleiotropy and enrichment of functional annotation . Statistical inference of the model parameters and SNP ranking is achieved through an EM algorithm that can handle genome-wide markers efficiently . When we applied GPA to jointly analyze five psychiatric disorders with annotation information , not only did GPA identify many weak signals missed by the traditional single phenotype analysis , but it also revealed relationships in the genetic architecture of these disorders . Using our hypothesis testing framework , statistically significant pleiotropic effects were detected among these psychiatric disorders , and the markers annotated in the central nervous system genes and eQTLs from the Genotype-Tissue Expression ( GTEx ) database were significantly enriched . We also applied GPA to a bladder cancer GWAS data set with the ENCODE DNase-seq data from 125 cell lines . GPA was able to detect cell lines that are biologically more relevant to bladder cancer . The R implementation of GPA is currently available at http://dongjunchung . github . io/GPA/ .
Hundreds of genome-wide association studies ( GWAS ) have been conducted to study the genetic bases of complex human traits . As of January , 2014 , more than 12 , 000 single-nucleotide polymorphisms ( SNPs ) have been reported to be significantly associated with at least one complex trait ( see the web resource of GWAS catalog [1] http://www . genome . gov/gwastudies/ ) . Despite of these successes , these significantly associated SNPs can only explain a small portion of genetic contributions to complex traits/diseases [2] . For example , human height is a highly heritable trait whose heritability is estimated to be around 80% , i . e . , 80% of variation in height within the same population can be attributed to genetic effects [3] . Based on large-scale GWAS , about 180 SNPs have been reported to be significantly associated with human height [4] . However , these loci together only explain about 5-10% of variation in height [2] , [4] , [5] . This phenomenon is referred to as the “missing heritability” [2] , [6] , [7] . Identifying the source of this missing heritability has drawn much attention from researchers , and progress has been made towards explaining the apparent discrepancy . The role of a much greater-than-expected set of common variants ( minor allele frequency ( MAF ) 0 . 01 ) has been shown to be critical in explaining the phenotypic variance [8] . Instead of only using genome-wide significant SNPs , Yang et al . [9] reported that , by using all genotyped common SNPs , 45% of the variance for human height can be explained . This result suggests that a large proportion of the heritability is not actually missing: given the limited sample size , many individual effects of genetic markers are too weak to pass the genome-wide significance , and thus those variants remain undiscovered . So far , people have found similar genetic architectures for many other complex traits [10] , such as metabolic syndrome traits [11] and psychiatric disorders [12]–[14] . That is , the phenotype is affected by many genetic variants with small or modest effects effects that cannot be confirmed individually via statistical significance , which is usually referred to as “polygenicity” . The polygenicity of complex traits is further supported by recent GWAS with larger sample sizes , in which more associated common SNPs with moderate effects have been identified ( e . g . , [15] ) . Clearly , the emerging polygenic genetic architecture imposes a great challenge of identifying risk genetic variants: a larger sample size is required to identify genetic variants with smaller effect sizes . However , sample recruitment may be expensive and time-consuming . It would be desirable to find a way to increase power to detect variants that miss significance on standard GWAS without extensive additional subject recruitment requirements . Integrative analysis of genomic data could be a promising direction , including combining GWAS data of multiple genetically related phenotypes and incorporating relevant biological information . The last few years have seen concrete demonstrations of “pleiotropy” , i . e . the sharing of genetic factors , between human complex traits . For example , a systematic analysis of the NHGRI catalog of published GWAS ( http://www . genome . gov/gwastudies/ ) showed that 16 . 9% of the reported genes and 4 . 6% of the reported SNPs are associated with multiple traits [16] . Through a “pleiotropic enrichment” method , Andreassen et al showed that it is possible to improve the power to detect schizophrenia-associated genetic variants by using the pleiotropy between schizophrenia ( SCZ ) and cardiovascular-disease [17] . A more recent study identified four significant loci ( p-value ) to have pleiotropic effects by analyzing GWAS data of 33 , 332 cases and 27 , 888 controls for five psychiatric disorders [18] . Further analysis suggested a very significant genetic correlation between schizophrenia and bipolar disorder ( s . e . ) [12] . Pleiotropy has also been demonstrated among several other types of traits , for example , metabolic syndrome traits [11] and cancers [19] . An increasing number of studies also suggest that functionally annotated SNPs are generally more biologically important than those that are not annotated , and henceforth more likely to be associated with complex traits . To name a few , using GWAS data of different traits ( e . g . , Crohn's disease and SCZ ) , Schork et al . [20] demonstrated a consistent pattern of enrichment of GWAS signals among functionally annotated SNPs . Yang et al . [21] showed that SNPs in genic regions could explain more variance of height and body mass index ( BMI ) than SNPs in intergenic regions . Nicolae et al . [22] found that SNPs associated with complex traits were more likely to be expression quantitative trait loci ( eQTL ) . In addition , public availability of a vast amount of functional annotation data also provides unprecedented opportunities to investigate the enrichment of GWAS signals among these various types of functional annotations . For example , the Encyclopedia of DNA Elements ( ENCODE ) Consortium has recently generated extensive experimental data on gene expression ( RNA-seq ) , DNA methylation status ( RRBS-seq ) , chromatin modifications ( ChIP-seq ) , chromatin accessibility ( DNase-seq and FAIRE-seq ) , transcription factor ( TF ) binding sites ( ChIP-seq ) , and long-range chromatin interactions ( ChIA-PET , Hi-C , and 5C ) . As of September 2012 , more than 1 , 600 data sets from 147 cell lines had been produced to annotate the human genome , including 2 . 89 million unique , non-overlapping DNase I hypersensitivity sites ( DHSs ) in 125 cell lines using DNase-seq and 630K binding regions of 119 DNA-binding proteins in 72 cell lines using ChIP-seq , among many [23] . The ENCODE Consortium [23] examined 4 , 492 risk-associated SNPs from the NHGRI GWAS catalog and found that 12% of them overlap with TF binding regions and 34% overlap with DHSs . The increasing evidence of pleiotropy between complex traits and the increasing functional annotation data call for novel statistical methods to effectively analyze multiple GWAS data sets and functional annotation data simultaneously . Statistical methods to investigate pleiotropy have been actively researched ( reviewed in [24] and [25] ) , for example , using linear mixed models [26] , [27] or the conditional False Discovery Rate ( FDR ) approach [17] , [28] . However , these methods do not allow use of functional annotation data for prioritization of GWAS results . On the other hand , various statistical methods have been proposed to make use of functionally annotated SNPs in recent years ( reviewed in [29] , [30] , and [31] ) . For example , GSEA [32] identifies potentially important pathways in which target genes of risk-associated SNPs are involved while RegulomeDB [33] allows nucleotide-level annotations of risk-associated SNPs , especially for those located in non-coding regions . Stratified FDR methods have been applied to incorporate annotation into GWAS data analysis [20] . However , each of these methods was designed for the analysis of a single phenotype and hence , they do not use functional annotation data fully efficiently for the genetic variants shared by multiple phenotypes . There is a need to develop a coherent statistical framework for the integration of functional annotation data for joint analysis of genetically correlated GWAS information . In this article , we propose a unified statistical framework , named GPA ( Genetic analysis incorporating Pleiotropy and Annotation ) , to prioritize GWAS results based on pleiotropy and annotation information . GPA also provides statistically rigorous and biologically interpretable inference tools for this purpose . The method can easily be used for other purposes as well , as we will discuss below . This article is organized as follows . First , we investigate the properties of GPA using extensive simulation studies and illustrate the versatility and utility of GPA with the analysis of real data . Specifically , we apply GPA to the five psychiatric disorder GWAS data from the Psychiatric Genomics Consortium with central nervous system gene expression data and show that GPA can accurately identify pleiotropy structure among these diseases . We further apply GPA to the bladder cancer GWAS data with the ENCODE DNase-seq data from 125 cell lines and show that GPA can detect cell lines that are biologically more relevant to bladder cancer . Lastly , we discuss many issues related to GPA . The details of our GPA model and its statistical inference procedures are provided in the Materials and Methods section .
We conducted comprehensive simulation studies to evaluate GPA performance . The p-values for non-risk SNPs can be simulated easily from a uniform distribution . For risk-SNPs , we can simulate their p-values via different approaches . The most favorable simulation for our GPA model is to simulate them from the Beta distribution . To examine the robustness of our GPA model , we adopted an alternative simulation scheme under the framework of the linear mixed model and liability threshold model that has gained increasing interest recently ( e . g . , [9] , [13] ) . The detailed procedures will be described later . But we emphasize that there is substantial discrepancy between the generative model used in simulation and the GPA model . The primary purpose of our simulation study is to investigate whether the GPA model can robustly improve the power to detect risk SNPs by integrating multiple GWAS data sets and annotation data despite this discrepancy . To simulate case-control GWAS data for two genetically correlated diseases , we followed the classical liability threshold model [13] . For each disease , we first simulated a large cohort of individuals with genotypes of independent SNPs . The MAFs of these SNPs were drawn uniformly from [0 . 05 , 0 . 5] . Then we randomly designated SNPs as risk SNPs . The per-minor-allele effect of each risk SNP was drawn from a normal distribution with zero-mean and variance of , where is the desired level of variance explained by all SNPs on the liability scale and is the MAF of the corresponding risk SNP . We also simulated the environmental effect on the liability scale for each individual from a standard normal distribution ( zero mean and unit variance ) . The total liability for each individual was then obtained by adding up all the genetic effects and the environmental effect . Given a desired disease prevalence , individuals with liabilities greater than the quantile were classified as cases and others were classified as controls . Then equal numbers of cases and controls were drawn from the cohort as a GWAS data set . When simulating two diseases simultaneously , we simulated two disjoint cohorts with the same set of SNPs . To reflect the pleiotropy effects between the two diseases , risk SNPs ( ) were chosen to be shared by the two diseases . The annotation status of each risk and non-risk SNP was simulated from a Bernoulli distribution with probability of and , respectively . In our simulation study , the total number of SNPs , , was set to be 20 , 000 , and the sample size of each data set , , was set at 2000 , 5000 or 10000 , respectively . The number of risk SNPs was the same for the two diseases and was set at 500 , 1000 or 2000 , respectively . We varied the proportion of shared risk SNPs between the two diseases , , from to 1 . Note that corresponds to the absence of pleiotropy . The disease prevalence , , was fixed at 0 . 1 and the variance explained by all the SNPs , , was fixed at 0 . 6 for each of the two diseases . Here and were fixed at 0 . 4 and 0 . 1 , respectively . We first evaluated SNP prioritization performance of GPA . Specifically , after the two data sets were simulated , we obtained the p-value for each SNP in each disease using a test with one degree of freedom . Then we analyzed the simulated data using our GPA method in the following four modes: 1 . analyzing the two diseases separately without the annotation data; 2 . analyzing the two diseases separately with the annotation data; 3 . analyzing the two diseases jointly without the annotation data; and 4 . analyzing the two diseases jointly with the annotation data . In each mode , we compared the order of the local FDR obtained using GPA against the actual risk status of the SNPs to calculate the area under the receiver operating characteristic curve ( AUC ) as a measure of risk SNP prioritization accuracy . The left panel of Figure 1 shows the AUCs from the four modes with and ( results for other scenarios are shown in Figures S10-S20 in Text S1 ) . Because all the simulation parameters were the same for the two diseases , only the results for the first disease are shown . We can see that incorporating either annotation information or pleiotropy between the two diseases improved the prioritization performance . In particular , as the proportion of shared risk SNPs increased , the prioritization performance also improved . Given the local FDR obtained using GPA , we controlled the global FDR at 0 . 2 and calculated the average power to identify the true risk SNPs . GPA performance measured by partial AUC and power for and are shown in the middle and right panels of Figure 1 , respectively ( results for other scenarios are shown in Figures S10–S20 in Text S1 ) . We also evaluated the actual FDR and found that the FDR was indeed controlled at 0 . 2 with occasional slight conservativeness ( Figure 2 and Figures S21–S31 in Text S1 ) . In addition , we evaluated the actual FDR in the presence of linkage disequilibrium ( LD ) among SNPs . The details of the simulations and results are given in Figure S1 in Text S1 . These results suggest that GPA can improve the power of identifying risk variants while controling FDR at the nominal level despite the mismatch between GPA and the generative model in our simulation study . Regarding parameter estimation , we found that GPA provided a satisfactory estimate of , the probability of being annotated for a certain group of SNPs , as long as there are enough SNPs in that group ( Figures S32–S44 in Text S1 ) . But we note that the estimates of the proportion of risk and non-risk SNPs , , may be biased ( Figures S45–S66 in Text S1 ) . This is no surprise because the distribution of the p-values of the risk SNPs obtained from the generative model in the simulation study may differ from the Beta distribution assumed in GPA . If the p-values of the risk SNPs are indeed generated from the Beta distribution , our GPA model can give fairly accurate estimates of ( Figures S67-S72 in Text S1 ) . As a comparison , we also used the “conditional FDR” approach proposed by Andreassen et al . [17] to prioritize SNPs in our simulations . The comparison results between GPA and the conditional FDR at and are shown in Figure 3 ( other results are provided in Figures S77–S87 in Text S1 ) . GPA significantly outperformed the conditional FDR approach in SNP prioritization . More importantly , in the absence of pleiotropy , the conditional FDR approach had worse accuracy than single-GWAS analysis using the standard FDR approach , whereas GPA achieved comparable accuracy with single-GWAS analysis in this scenario . This suggests that GPA was able to take advantage of pleiotropy when it exists while , in clear contrast to the conditional FDR approach , it does not sacrifice much statistical power than the conditional FDR approach when it is absent . Next , we evaluated the type I error and power of GPA for hypothesis testing on the significance of annotation enrichment for risk SNPs . Gene Set Enrichment Analysis ( GSEA ) [34] is a popular method to accomplish a similar task . Although GSEA typically is used for gene expression data analysis , its input can be a list of p-values obtained from any source . Therefore we implemented the GSEA method to test the enrichment of the -values of a set of SNPs being annotated and compared it with GPA . We followed the previous simulation scheme and simulated one GWAS data set with , varying from 2000 to 10000 , and varying from 500 to 2000 . Here was fixed at 0 . 1 and was varied from 0 . 1 to 0 . 5 . We set the statistical significance level at 0 . 05 . Type I error rate was evaluated at and power was evaluated at . The results for are shown in Figure 4 . In general , GPA provided much higher power than GSEA while both methods appropriately controlled the type I error rate . We further evaluated the type I error rate and power of GPA for the test of pleiotropy in our simulations . The simulation parameters were the same as those in the previous simulations . Power was evaluated at , , , and . The type I error rate was evaluated at , corresponding to the expected shared proportion of risk SNPs in the absence of pleiotropy . As shown in Figure 5 , power increased as decreased and as and increased , whereas the type I error rate was appropriately controlled in all cases . Please note that type I errors and power remained almost the same for hypothesis testing of pleiotropy in the presence of annotation ( see Figures S2–S4 in Text S1 ) . Lastly , we performed additional simulations with moderate heritability ( ) and pleiotropy ( ) . The results shown in Figures S73–S76 in Text S1 demonstrate that , with moderate heritability and pleiotropy , GPA can still effectively improve the power by leveraging pleiotropy between related traits . Therefore , GPA can serve as a more powerful tool for integrative analysis in the post-GWAS era . DNase I hypersensitive sites ( DHSs ) are regions where DNA degradation by enzymes like DNase I occur more frequently than elsewhere . As a result , DHSs can mark active transcription regions across genome and these patterns are known to be tissue or cell specific . The ENCODE project analyzed the DHSs in 125 human cell lines with the intention of cataloging human regulatory DNA [38] . In this section , we applied GPA to assess how bladder cancer [39] risk associated SNPs are enriched in DHSs region across these 125 human cell lines . We downloaded genotype data for bladder cancer from dbGaP ( NCI Cancer Genetic Markers of Susceptibility ( CGEMS ) project; accession number phs000346 . v1 . p1 ) . We used samples genotyped from both Illumina 1 M chip and 610K chip for our analysis . For quality control , we removed SNPs with missing rates 0 . 01 . We checked Hardy-Weinberg Equilibrium and excluded SNPs with p-value 0 . 001 . SNPs with minor allele frequencies ( MAF ) were also removed . After quality control , 490 , 614 SNPs from 3 , 631 cases and 3 , 356 controls of European descent were used in the analysis . SNP-level association p-values were calculated for this bladder cancer data using logistic regression by assuming an additive genetic model . We also downloaded the uniform peak files for DHSs in 125 cell lines from the ENCODE project ( http://genome . ucsc . edu/cgi-bin/hgFileUi ? db=hg19&g=wgEncodeAwgDnaseUniform ) . Note that the DHSs for these 125 cell lines were identified with a uniform analysis workflow by the ENCODE Consortium; this facilitates fair and unbiased comparison among cell lines as annotation for our GPA model . We applied GPA to analyze the bladder cancer GWAS -values with one annotation dataset at a time , and performed hypothesis testing to assess the significance of enrichment . The results are shown in the left panel of Figure 8 . Under significance level after Bonferroni correction , annotations from 19 cell lines were statistically significantly enriched for bladder cancer risk associated SNPs . Most of these cell lines were derived from lymphocytes from normal blood ( e . g . , T cells CD4+ Th0 adult , Monocytes CD14+ RO01746 ) , while some cell lines came from cancer patients ( e . g . , Gliobla and HeLa-S3 ) . The above results suggest that involvement of the immune system and carcinoma pathways in bladder cancer . These results also demonstrate that GPA may be an effective way to explore functional roles of GWAS hits by testing enrichment on phenotype-related annotations or user-specified annotations . We also compared GPA with the LMM-based approach [21] , [40] for this dataset . Specifically , we considered the following genome-partitioning LMM: ( 1 ) where are covariates ( the first five principal components from genotype data ) , and and are sets of SNPs overlapping DHSs in each cell line and the remaining SNPs , respectively . We denote the numbers of SNPs in and as and , respectively . The median number of SNPs that overlap DHS in each cell line is about 60K and 90% of cell lines have the number of DHSs ranging between 40K and 80K . In order to take into account such variation in DHS number among cell lines , we define a scaled version of the proportion of phenotype variance explained by SNPs overlapping DHSs in each cell line as ( 2 ) where is the proportion of the explained variance and is the scaling factor . The right panel of Figure 8 shows that the ( ) -transformed p-value of the GPA annotation enrichment test is linearly related to . This indicates that our GPA model captures enrichment of annotation almost as accurately as LMM even without the original genotype data , implying its broader applicability than methods requiring individual genotype and phenotype data .
Many GWAS have been conducted in the past 10 years that have led to the discoveries of thousands of genomic regions associated with many traits , and many more discoveries are expected from ongoing GWAS . As GWAS data accumulate , there is an urgent need to perform a systematic analysis of available GWAS datasets for a comprehensive understanding of the genetic architecture of complex traits , and provide new insights for functional roles of the implicated variants . To achieve this goal , there is a great interest in developing computational and statistical approaches to exploring genomic data in the post-GWAS era . In the following , we briefly discuss the relationship between GPA and other related methods , such as LMM , conditional FDR and GSEA . LMM is an effective method for exploring genetic architecture of complex traits and it has been implemented in a popular software package , GCTA [41] . Compared with LMM , GPA has the following distinctive features: To our best knowledge , the conditional FDR approach is the first approach that statistically addresses the issue of pleiotropy between two GWAS . and GSEA is presently the most popular approach to evaluating the enrichment of gene sets . In fact , GPA provides a unified framework for systematically integrating both sources of information , pleiotropy and annotation . Rigorous statistical inference of pleiotropic effects and annotation enrichment has been established in this framework . As demonstrated in our extensive simulation study , GPA has better performance of identifying disease-associated markers than the conditional FDR approach , and it shows greater power of evaluating annotation enrichment than GSEA as well . With real data analysis , we have demonstrated how to use GPA to incorporate pleiotropy information and multiple annotation data for prioritizing GWAS results . Here we briefly discuss some key assumptions made in GPA . The parameters in the GPA model should be interpreted with caution because parameter estimation is based on the model assumption as discussed above . For example , as we showed in simulation study , the estimated can be biased due to the mismatch of GPA model and the random-effects model . There are also some limitations of the current GPA model . Although extensions to three or more GWAS are straightforward in principle ( from four-groups model to -groups model , ) , the number of groups will increase exponentially as the number of GWAS increases . This makes many close to zero , resulting in poor parameter estimation ( large variance ) and thus reduced power . Currently , we suggest doing pairwise-GWAS analysis to explore the genetic relationship of different phenotypes . It is of great interest to develop statistical models to handle the multiple-GWAS case more effectively . Another limitation is that current GPA version can only handle binary annotation . Allowing more annotation structures ( e . g . , continuous annotation ) in GPA is an important extension of current GPA model . We will investigate this issue in the future . In summary , we have presented a statistical approach , named GPA , that can integrate information from multiple GWAS data sets and functional annotation data . Not only does GPA have better statistical power than related methods , it also provides interpretable model parameters offering insights to our understanding of the genetic architecture of complex traits . We have successfully applied GPA to analyze GWAS data of five psychiatric disorders from PGC , and showed that GPA is able to identify pleiotropic effects among psychiatric disorders and detect enrichment of the CNS gene set . We have also applied GPA to analyze a bladder cancer GWAS dataset with ENCODE data as annotation , where significant enrichments of immune system and carcinoma pathways were observed . Compared to LMM that requires individual genotype and phenotype data as input , GPA has similar results of enrichment analysis without requirements of the genotype data . This makes GPA an attractive and effective tool for the integrative analysis of multiple GWAS data with functional annotation data , when genotype data are not available .
Throughout this paper , we shall use to index SNPs , to index GWAS data sets , and to index the annotation data sets . We first consider the simplest case where we only have summary statistics ( p-values ) from just one GWAS data set , and then extend our model to handle multiple GWAS data sets and annotation data . Suppose we have performed hypothesis testing of genome-wide SNPs and obtained their p-values: ( 3 ) where is the number of SNPs . Consider the “two-groups model” [46] , i . e . , the obtained p-values are assumed to come from the mixture of null and non-null , with probability and , respectively . Let be the latent variables indicating whether the j-th SNP is null or non-null , where , , and , because a SNP can only be either null or non-null . Here means un-associated ( null ) and means associated ( non-null ) . Then we have the following two-groups model: ( 4 ) where the p-values from the null group are from the Uniform distribution on [0 , 1] , denoted as , and the p-values from the non-null group are from the Beta distribution with parameters ( ) , where . We put the constraint to model that a smaller p-value is more likely than a larger p-value when it is from the non-null group [47] . To incorporate information from functional annotation data , we extend the basic model as follows . Suppose we have collected information from functional annotation sources in the annotation matrix: , where indicates whether the j-th SNP is annotated in the d-th functional annotation source . For example , when there are two annotation sources – eQTL data and DNase I hypersensitivity sites ( DHS ) data – then is an matrix . If the j-th SNP is an eQTL , then , otherwise ; if it is located in a DHS , then , otherwise . Now we model the relationship between and as ( 5 ) Clearly , can be interpreted as the proportion of null SNPs being annotated in the d-th annotation , and corresponds to the proportion of non-null SNPs being annotated in the d-th annotation . Therefore , implies that there exists enrichment for the d-th annotation . The statistical inference about enrichment of annotation data will be discussed in details in Section “Hypothesis testing of annotation enrichment and pleiotropy” . Now we extend the above model to handle multiple GWAS data sets . To keep the notation uncluttered , we present the model for the case of two GWAS data sets . Suppose we have p-values from two GWAS: ( 6 ) Let be the matrix collecting all the p-values , where denotes the p-value of the j-th SNP in the k-th GWAS . Similarly , we introduce latent variables indicating the association between the j-th SNP and the two phenotypes: means the j-th SNP is associated with neither of them , means it is only associated with the first one , means it is only associated with the second one , and means it is associated with both . The two-groups model ( 4 ) is extended to the following “four-groups model”: ( 7 ) where . When the genetic bases of the two phenotypes are independent of each other ( i . e . , no pleiotropy ) , then we have by expectation . Therefore , the difference between and can be used to characterize pleiotropy . Statistical inference on pleiotropy is given in Section “Hypothesis testing of annotation enrichment and pleiotropy” . To incorporate annotation information into the multiple GWAS model ( 7 ) , similarly , we model the relationship between and as ( 8 ) where is the probability of a null SNP being annotated , is the probability of the first phenotype associated-SNP being annotated , is the probability of the second phenotype associated-SNP being annotated , and is the probability of jointly associated-SNP being annotated . Assuming the independence of SNP markers , the joint distribution can be written as ( 9 ) where and are the j-th row of and ; the second equation holds by assuming the independence between and , conditional on ; and the third equation holds by further assuming the independence between and for , conditional on . Parameters in the GPA model can be estimated using the Expectation-Maximization ( EM ) algorithm [48] , which is computationally efficient because we have explicit solutions for estimation of all the parameters in the M-step . Standard errors for parameter estimates can be approximated using the empirical observed information matrix [49] . Note that in the GPA model , the sample size for estimating the empirical observed information matrix corresponds to the number of SNPs and as a result , we have a very large sample size ( ) to estimate standard errors accurately . More details of the EM algorithm and the estimation of standard errors are provided in Sections 1 and 3 in Text S1 . | In the past 10 years , many genome wide association studies ( GWAS ) have been conducted to identify the genetic bases of complex human traits . As of January , 2014 , more than 12 , 000 single-nucleotide polymorphisms ( SNPs ) have been reported to be significantly associated with at least one complex trait/disease . On one hand , about 85% of identified risk variants are located in non-coding regions , which motivates a systematic understanding of the function of non-coding variants in regulatory elements in the human genome . On the other hand , complex diseases are often affected by many genetic variants with small or moderate effects . To address these issues , we propose a statistical approach , GPA , to integrating information from multiple GWAS datasets and functional annotation . Notably , our approach only requires marker-wise p-values as input , making it especially useful when only summary statistics , instead of the full genotype and phenotype data , are available . We applied GPA to analyze GWAS datasets of five psychiatric disorders and bladder cancer , where the central nervous system genes , eQTLs from the Genotype-Tissue Expression ( GTEx ) , and the ENCODE DNase-seq data from 125 cell lines were used as functional annotation . The analysis results suggest that GPA is an effective method for integrative data analysis in the post-GWAS era . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"genome-wide",
"association",
"studies",
"mathematics",
"statistics",
"(mathematics)",
"genome",
"analysis",
"genetics",
"biology",
"and",
"life",
"sciences",
"physical",
"sciences",
"computational",
"biology",
"human",
"genetics"
] | 2014 | GPA: A Statistical Approach to Prioritizing GWAS Results by Integrating Pleiotropy and Annotation |
Chikungunya virus ( CHIKV ) is an emerging arbovirus associated with several recent large-scale epidemics . The 2005–2006 epidemic on Reunion island that resulted in approximately 266 , 000 human cases was associated with a strain of CHIKV with a mutation in the envelope protein gene ( E1-A226V ) . To test the hypothesis that this mutation in the epidemic CHIKV ( strain LR2006 OPY1 ) might influence fitness for different vector species , viral infectivity , dissemination , and transmission of CHIKV were compared in Aedes albopictus , the species implicated in the epidemic , and the recognized vector Ae . aegypti . Using viral infectious clones of the Reunion strain and a West African strain of CHIKV , into which either the E1–226 A or V mutation was engineered , we demonstrated that the E1-A226V mutation was directly responsible for a significant increase in CHIKV infectivity for Ae . albopictus , and led to more efficient viral dissemination into mosquito secondary organs and transmission to suckling mice . This mutation caused a marginal decrease in CHIKV Ae . aegypti midgut infectivity , had no effect on viral dissemination , and was associated with a slight increase in transmission by Ae . aegypti to suckling mice in competition experiments . The effect of the E1-A226V mutation on cholesterol dependence of CHIKV was also analyzed , revealing an association between cholesterol dependence and increased fitness of CHIKV in Ae . albopictus . Our observation that a single amino acid substitution can influence vector specificity provides a plausible explanation of how this mutant virus caused an epidemic in a region lacking the typical vector . This has important implications with respect to how viruses may establish a transmission cycle when introduced into a new area . Due to the widespread distribution of Ae . albopictus , this mutation increases the potential for CHIKV to permanently extend its range into Europe and the Americas .
The large-scale epidemic of the mosquito-transmitted alphavirus , Chikungunya virus ( CHIKV ) , began in Kenya in 2004 and spread to several Indian Ocean islands including the Comoros , Mauritius , the Seychelles , Madagascar , Mayotte and Reunion . On Reunion island alone there were approximately 266 , 000 cases ( 34% of the total island population ) [1–6] . In the continuing Indian epidemic there have been at least 1 . 4M cases reported [7–10] with continued expansion in Sri Lanka and Indonesia . CHIKV had not been reported to cause fatalities in prior outbreaks; however , during the outbreak on Reunion island , CHIKV was associated with at least 260 deaths [11 , 12] . The strain of CHIKV responsible for the Indian Ocean island epidemic has been well-characterized in cell culture and mosquito models [13–15]; however , the underlying genetic basis of the atypical phenotype of this CHIKV strain remains unknown . CHIKV is transmitted by Aedes species mosquitoes , primarily Ae . aegypti . However , the 2005–2006 CHIKV epidemic on Reunion island was unusual because the vector responsible for transmission between humans was apparently the Asian tiger mosquito , Ae . albopictus [3 , 16] . This conclusion is based on several factors . This species is known to be susceptible to CHIKV infection and although infectious virus was not isolated from Ae . albopictus during the epidemic , CHIKV RNA was detected ( X . de Lamballerie , personal communication ) . Furthermore , the species is anthropophylic , was abundant during the epidemic , and other potential vectors specifically Ae . aegypti were relatively scarce with a very limited distribution ( P . Reiter , personal communication ) . Ae . albopictus is abundant and widely distributed in urban areas of Europe and the United States of America [17–22] . CHIKV infections have been reported in many travelers returning to the US and Europe [12 , 23–26] causing concern that the virus could be introduced and become established in these areas [1 , 27 , 28] . In August and September of 2007 , a CHIKV–Ae . albopictus transmission cycle was reported for the first time in Europe , with an estimated 254 human cases occurring in Italy [29 , 30] . Alphaviruses are enveloped single stranded positive sense RNA viruses . Genomic RNA , of ≈ 12 , 000 nt , encodes four non-structural ( ns1–4 ) and three main structural proteins ( capsid , E2 and E1 ) . At neutral pH , E2 and E1 exist as heterodimers in which E2 forms spikes on the virion surface that interact with cellular receptors . The E1 protein lies below E2 and mediates fusion of the viral and cellular membranes during viral entry [31] . Analysis of CHIKV genome microevolution during the 2005–2006 Indian Ocean epidemic identified an alanine to valine mutation at position 226 in the E1 envelope glycoprotein ( E1-A226V ) among viral isolates obtained during the outbreak [32] . The reason for this was unclear but it was hypothesized that the E1-A226V mutation might influence infectivity of CHIKV for mosquito vectors [11 , 32] . Interestingly , earlier studies have identified that a P→S mutation in the same position of the E1 glycoprotein is responsible for the modulation of Semliki Forest virus's ( SFV , a member of the alphavirus family ) requirements for cholesterol in the target membrane [33] . It also has been shown that the presence of this mutation results in more efficient growth of SFV in Ae . albopictus mosquitoes [34] . However , no evidence has been presented to directly correlate the release from the cholesterol dependence , associated with the E1-P226S mutation in SFV , with a growth advantage in Ae . albopictus . It is unknown if dependence on cholesterol for growth in mosquito cells is a requirement of all alphaviruses . To test the hypothesis that the E1-A226V mutation might influence the fitness of CHIKV in mosquito vectors , we compared the effect of this mutation on CHIKV mosquito infectivity , the ability to disseminate into heads and salivary glands , and the relative fitness in competition assays for transmission by Ae . albopictus and Ae . aegypti to suckling mice . We also analyzed the effect of the E1-A226V mutation on CHIKV cholesterol dependence for growth in mosquito C6/36 ( Ae . albopictus ) cells . Here we report findings that a single nucleotide change , which arose during the epidemic , significantly increases fitness of the virus for Ae . albopictus mosquitoes and was associated with CHIKV dependence on cholesterol in the mosquito cell membrane . This change likely enhanced CHIKV transmission by an atypical vector and contributed to the maintenance and scale of the epidemic .
To test the hypothesis that the E1-A226V mutation altered CHIKV infectivity for Ae . albopictus mosquitoes , CHIKV infectious clones derived from an epidemic Reunion island human isolate were used [15] , including one clone ( LR-GFP-226V ) expressing enhanced green fluorescent protein ( eGFP ) . Clones were further engineered to express E1 protein containing an alanine at position E1–226 ( LR-GFP-226A ) representing the CHIKV genotype prevalent prior to the outbreak gaining momentum ( Figure S1 ) . RNAs produced from both clones ( LR-GFP-226V and LR-GFP-226A ) have comparable specific infectivity values , produced similar viral titers following transfection into BHK-21 cells ( Table S1 ) and have similar growth kinetics in mosquito ( C6/36 ) and mammalian ( BHK-21 ) cells lines ( Figure S2A and S2B ) . The relative infectivity of LR-GFP-226V and LR-GFP-226A viruses was analyzed in female Ae . albopictus mosquitoes orally exposed to serial 10-fold dilutions of CHIKV ( LR-GFP-226 V or A ) . To determine whether infection rates correlate with blood meal titer , midguts dissected from mosquitoes at 7 days post-infection ( dpi ) were analyzed for foci of eGFP-expressing cells by fluorescence microscopy ( Figure 1A; Table 1 ) . In two independent experiments , LR-GFP-226V virus was found to be approximately 100-fold more infectious to Ae . albopictus than LR-GFP-226A virus ( p<0 . 01 ) . To test if the infectivity phenotype was directly linked to the mutation , the complementary reverse mutation , E1-A226V , was introduced into an infectious clone of a West African CHIKV strain , 37997-GFP ( 37997-GFP-226A ) ( Figure S1 ) . The Reunion and 37997 strains of CHIKV are distantly related , with only 85% nucleotide sequence identity . The parental 37997-GFP-226A and the 37997-GFP-226V viruses were indistinguishable in cell culture experiments ( Table S1; Figure S2C and S2D ) ; however , in vivo experiments in Ae . albopictus mosquitoes revealed that the E1-A226V mutation significantly decreases the oral infectious dose 50 ( OID50 ) value for the 37997-GFP-226V virus ( p<0 . 01 ) to an extent similar to that observed for LR-GFP-226V virus ( Figure 1B; Table 1 ) . These data conclusively demonstrate that the single E1-A226V point mutation is therefore sufficient to significantly reduce the OID50 of the 37997-GFP virus ( p<0 . 01 ) in Ae . albopictus mosquitoes equivalent to that observed for the LR-GFP-226V virus ( Figure 1A; Table 1 ) . To further evaluate viral fitness of the epidemic CHIKV E1-A226V mutation in Ae . albopictus , viral competition experiments were performed . Although our CHIKV eGFP-expressing infectious clones , have similar infection properties in mosquitoes as wild-type viruses [15 , 35] , to address potential concerns that eGFP expression might influence OID50 values , we constructed LR-226A and LR-ApaI-226V viruses without eGFP and employed them in viral competition experiments ( Figures 2A and S1 ) . LR-ApaI-226V was derived from previously described CHIK-LR ic , by the introduction of a silent marker mutation , A6454C , in order to add an ApaI restriction site into the coding sequence . It was shown that the A6454C mutation does not affect the specific infectivity value ( Table S1 ) , the viral titer after RNA transfection into BHK-21 cells value ( Table S1 ) , the viral growth kinetics in BHK-21 and C6/36 cells ( Figure S3 ) , infectivity for and viral titers in Ae . aegypti and Ae . albopictus mosquitoes ( Table S2 ) , or viral fitness for growth in BHK-21 and C6/36 cells as determined by competition assay ( Figure S4 ) . These data indicate that the introduced mutation is indeed silent and does not affect the fitness of LR-ApaI-226V . For viral competition experiments LR-ApaI-226V virus ( 107 plaque-forming units ( pfu ) ) was mixed with an equal amount of LR-226A virus . LR-ApaI-226V and LR-226A viruses are indistinguishable in cell culture experiments ( Figure S3 ) . Mixtures of LR-ApaI-226V and LR-226A viruses were orally presented to Ae . albopictus mosquitoes in a blood meal , and midguts were examined at 7 dpi . The relative amount of RNA derived from LR-ApaI-226V in the midgut cells increased 5 . 7±0 . 6 times as compared to the initial relative amount of LR-ApaI-226V RNA in the blood meal sample ( Figure 2B ) . These data support our observation that the E1-A226V mutation enhances infectivity of CHIKV for Ae . albopictus mosquitoes and furthermore demonstrate that the mutation could provide an evolutionary advantage over E1-226A viruses in an atypical vector and may have perpetuated the outbreak in a region where Ae . albopictus was the predominant anthropophilic mosquito species . To determine if the enhanced midgut infectivity associated with the E1-A226V mutation may result in more efficient viral dissemination into secondary tissues , the kinetics of viral dissemination by LR-GFP-226V and LR-GFP-226A into salivary glands , and competition between LR-ApaI-226V and LR-226A for dissemination into mosquito heads were analyzed ( Figure 3A and 3B ) . LR-GFP-226V virus disseminated more rapidly into Ae . albopictus salivary glands at all time points , with a significant difference at 7 dpi ( p=0 . 044 , Fisher's exact test ) . Similarly , in three of four replicates of competition experiments , RNA from LR-ApaI-226V virus was dramatically more abundant in the heads of Ae . albopictus mosquitoes as compared to RNA from LR-226A ( Figure 3B , lines 1 , 3 , 4 ) , although in one replica LR-ApaI-226V RNA was only slightly more abundant as compared to the initial viral RNA ratio ( Figure 3B , line 2 ) . This variability of the results may be due to random pooling of mosquito heads . Thus , replicate two may have included more heads negative for LR-Apal-226V relative to heads positive for LR-226A RNA . Another possibility is that at some point during viral dissemination from the midguts into mosquito heads , LR-226A may replicate more rapidly than LR-ApaI-226V . To further investigate this relationship , Ae . albopictus mosquitoes were orally presented with either LR-ApaI-226V or LR-226A and whole mosquito body viral titers were compared at different time points pi . Surprisingly , no significant differences between viral titers were found , with the exception of 1 dpi , where the LR-ApaI-226V titer was 0 . 5 Log10 tissues culture infectious dose 50 percent end point titer ( Log10 TCID50/mosquito ) higher than of the LR-226A titer ( Figure 4A ) . This may be due to more efficient colonization of Ae . albopictus midguts by LR-ApaI-226V . The absence of significant differences in viral titers at later time points may be due to variation in viral titers among individual mosquitoes . Competition between LR-ApaI-226V and LR-226A was analyzed at different time points in order to investigate the relationship between replication of LR-ApaI-226V and LR-226A viruses in Ae . albopictus mosquitoes ( Figure 4B ) . As expected , the viral RNA from LR-ApaI-226V was predominant at the early time points of 1 and 3 dpi . Interestingly , between 3 and 5 dpi the viral RNA ratio shifted toward LR-226A virus indicating that at these time points , LR-226A replicates more efficiently in some mosquito tissues ( Figure 4B ) . This short period of time may have a slight effect on the overall outcome of competition for dissemination into salivary glands because there is a reverse shift in the RNA ratio between days 5 and 7 toward LR-ApaI-226V virus , which continues through 14 dpi . These data indicate that the E1-A226V mutation not only increases midgut infectivity but also is associated with more efficient viral dissemination from the midgut into secondary organs , suggesting that the E1-A226V mutation would increase transmissibility of CHIKV by Ae . albopictus mosquitoes . A competition assay between LR-ApaI-226V and LR-226A viruses was used to examine transmission by Ae . albopictus to suckling mice to assess the potential for the E1-A226V mutation to influence virus transmission . Ae . albopictus mosquitoes were orally presented with a mixture of LR-ApaI-226V and LR-226A viruses and at 14 dpi were allowed to feed on suckling mice . Mice were sacrificed and bled on day 3 following exposure and the presence of CHIKV RNA in the blood was analyzed by RT-PCR followed by restriction digestion with ApaI ( Figure 5B ) . Blood obtained from 100% of experimental mice contained detectible amounts of viral RNA , indicating that virus was transmitted by Ae . albopictus mosquitoes to suckling mice . More importantly , in all six mice analyzed , RNA derived from LR-ApaI-226V was the predominant viral RNA species , indicating that under the conditions of competition for transmission , the E1-A226V mutation directly increases CHIKV transmission by Ae . albopictus mosquitoes . Interestingly , in the control experiment in which mice were subcutaneously inoculated with ≈ 50 pfu of 1:1 mixture of LR-ApaI-226V and LR-226A viruses , RNAs from both viruses were readily detected and no difference was observed in the viral RNA ratio 3 dpi ( Figure 5A ) indicating that at least in mice , E1-A226V is not associated with changes in viral fitness . Since the E1-A226V mutation confers a fitness advantage in Ae . albopictus , it is unknown why this mutation had not been observed previously . It is possible that this change might have a deleterious effect on viral fitness in the vertebrate host , although our data of direct competition of LR-ApaI-226V and LR-226A viruses in suckling mice ( Figure 5A ) and analysis of CHIKV cellular tropism of four clinical isolates from Reunion ( which have either A or V at position E1–226 ) [14] , suggest that this is unlikely . An alternative hypothesis is that the E1-A226V mutation might compromise the fitness of CHIKV or have neutral fitness effects in the mosquito species which served as a vector for CHIKV prior to its emergence on Reunion island . Since Ae . aegypti has generally been regarded as the main vector for CHIKV prior to the emergence on Reunion island , we analyzed the effect of the E1-A226V mutation on fitness of CHIKV in Ae . aegypti . In contrast to the results obtained in Ae . albopictus mosquitoes , OID50 values of viruses containing the E1-226V in the backbone of the Reunion and 37997 strains of CHIKV were approximately 0 . 5 Log10OID50/ml higher than the OID50 values of E1-226A viruses in all experiments using Ae . aegypti . These differences were statistically significant for one out of two replicates for each virus pair ( Figure 1C and 1D; Table 2 ) . A competition assay examining LR-ApaI-226V and LR-226A virus infection in Ae . aegypti midguts , demonstrated that LR-226A virus out-competed LR-ApaI-226V virus at 7 dpi in all four replicates using ten midguts per replicate and that the amount of LR-226A RNA increased on average 3 . 1 times as compared to the initial blood meal RNA ratio ( Figure 2C ) . These data suggest that the E1-A226V mutation has a slight negative effect on CHIKV infectivity of Ae . aegypti midguts . The effect of the E1-A226V mutation on the ability of CHIKV to disseminate into Ae . aegypti secondary organs was also analyzed ( Figure 3C and 3D ) . LR-GFP-226V and LR-GFP-226A viruses both have similar kinetics of dissemination into salivary glands following oral infection using titers 1–2 Log10TCID50 higher than their OID50 value in Ae . aegypti ( Figure 3C ) . In a competition assay , both LR-ApaI-226V and LR-226A viruses disseminated similarly into the heads of Ae . aegypti . In two of four replicas , there was a slight increase in the relative amount of LR- 226A RNA ( Figure 3D , lines 1 , 4 ) ; whereas the other two replicas showed a decrease in LR-226A RNA ( Figure 2D , lines 2 , 3 ) , relative to the initial ratio of the RNA of LR-ApaI-226V and LR-226A viruses in the blood meal . A competition of LR-ApaI-226V and LR-226A viruses for transmission by Ae . aegypti to suckling mice was also analyzed ( Figure 5C ) . In contrast to transmission by Ae . albopictus mosquitoes , five out of six mice fed upon by Ae . aegypti contained comparable amounts of RNA derived from both viruses and only one out of six mice contained RNA derived exclusively from LR-ApaI-226V . It has been previously shown that a P→S mutation in the same E1–226 position of SFV releases cholesterol dependence of the virus in C6/36 cells [33] and results in significantly more rapid growth of SFV in Ae . albopictus mosquitoes after intrathoracic inoculation [34] . To determine if a requirement for cholesterol in the cell membrane is important for CHIKV , we analyzed cholesterol dependence of CHIKV E1-226A and E1-226V viruses ( Figure 6 ) . Growth curves of E1-226A and E1-226V viruses in the background of Indian Ocean and West African strains of CHIKV were almost indistinguishable when grown in C6/36 cells maintained in L-15 supplied with standard 10% FBS ( Figure 6A ) . However , when the cells were depleted of cholesterol , LR-226A and 37997–226A viruses replicated significantly more rapidly than LR-226V and 37997–226V viruses , reaching 3 Log10TCID50/ml higher titer at 1 , 2 and 3 dpi ( Figure 6B ) . These data indicate that adaptation of CHIKV to Ae . albopictus mosquitoes coincides with CHIKV dependence on cholesterol in the target cell membrane .
The CHIKV outbreak in Reunion is unique because it is the first well-documented report of an alphavirus outbreak for which Ae . albopictus was the main vector . Interestingly , this was also the first Chikungunya epidemic during which fatal infections were reported . Our data clearly indicate that an E1-A226V mutation in CHIKV results in increased fitness of CHIKV in Ae . albopictus mosquitoes with respect to midgut infectivity , dissemination to the salivary glands , and transmission to a vertebrate species . These data demonstrate that a single E1-A226V mutation is sufficient to dramatically increase the ability of different strains of CHIKV to infect Ae . albopictus mosquitoes and that this substitution requires no additional adaptive mutations to gain intermolecular compatibility . These complimentary experimental data demonstrate that a single mutation is sufficient to modify viral infectivity for a specific vector species and as a consequence , can fuel an epidemic in a region that lacks the typical vector . These observations provide the basis for an explanation of the observed rapid shift among CHIKV genotypes to viruses containing the E1-A226V mutation during the Reunion outbreak [32] . Interestingly , our data and data from previous studies [36 , 37] indicate that prior to acquiring the E1-A226V mutation , CHIKV is capable of producing high enough viremia in humans to efficiently infect Ae . albopictus mosquitoes . One explanation of the evolutionary force which allowed CHIKV to be selected so rapidly into a CHIKV strain which is adapted to Ae . albopictus , is that the increased infectivity ( lower OID50 ) of CHIKV E1-A226V mutants for Ae . albopictus means that the human viremic thresholds required for Ae . albopictus infection would likely occur earlier and be sustained for longer . Several recent studies indicate that during the course of human viremia , which last up to 6 days , CHIKV loads can reach up to 3 . 3x109 RNA copies per ml of the blood [38 , 39] , which corresponds to 6–7 Log10TCID50/ml [39] . Earlier studies that utilized a suckling mouse brain titration protocol , which is more sensitive than titration on Vero cells , also found that human viremia often exceeded 6 Log10SMICLD50/0 . 02 ml [40] . Based on viremia studies in rhesus monkeys that can develop up to 7 . 5 Log/ml if assayed by suckling mice brain titration [41] and a maximum viremia of only 5 . 5 Log10/ml based on Vero cell titration [42] , we believe that viremias in humans would correlate to 6–7 Log10TCID50/ml . From these data we calculate that the maximum virus load which can be achieved in human blood is 1–2 Log10TCID50/ml higher than the Log10OID50/ml for E1-226A viruses but 3–4 Log10TCID50/ml higher than the Log10OID50/ml for E1-226V viruses . During the course of viremia there should therefore be a substantial time frame in which CHIKV blood load is high enough for E1-226V viruses to infect Ae . albopictus but below the threshold for infection with E1-226A viruses . This increased opportunity for Ae . albopictus infection , would perpetuate the selection and transmission of the mutant virus . During transmission competition assays , only E1-226V virus was transmitted to suckling mice by Ae . albopictus , although in these experiments , titers of E1-226V and E1-226A viruses were of a high enough magnitude to allow both of these viruses to efficiently infect this mosquitoes species . This indicates that there are additional mechanisms that could ensure evolutional success of the E1-A226V viruses transmitted by Ae . albopictus . It is possible that one of these mechanisms is associated with more efficient dissemination of the E1-226V as compared with E1-226A viruses . This could shorten the extrinsic incubation period ( EIP ) —the time from mosquito infection to transmission—and could have contributed to the evolutionary success of CHIKV during the Reunion outbreak because vectors infected with the LR-226V virus would transmit it more quickly than those infected with LR-226A viruses . Additionally , with relatively short-lived vectors such as mosquitoes [43] , longer EIPs reduce transmission efficiency simply because fewer mosquitoes survive long enough to transmit the virus . Our current studies do not provide data to determine if dissemination efficiency of the E1-226V viruses into the salivary glands is a consequence of more efficient midgut infectivity or if these two phenomena are independent . In this regard , it will be of particular interest to investigate the effect of the E1-A226V mutation on CHIKV transmission by orally or intrathoracically infected Ae . albopictus mosquitoes . Although the CHIKV E1-A226V mutation gives a selective advantage in Ae . albopictus , there was not a corresponding advantage in Ae . aegypti . The OID50 and midgut competition assay data indicate that E1-226V viruses were slightly less infectious for midgut cells of Ae . aegypti mosquitoes ( Figures 1C , 1D , and 2C; Table 2 ) . Additionally , in contrast to Ae . albopictus , E1-226V viruses do not have a detectable advantage for dissemination into salivary glands and heads of Ae . aegypti . In transmission competition experiments from Ae aegypti to suckling mice , E1-226V conferred a slight competitive advantage over E1-226A ( Figure 5C ) . However , five out of six mice exposed to CHIKV infected Ae aegypti had equivalent amounts of both E1-226A and E1-226V viral RNAs . These results are markedly different compared to the results obtained in similar experiments using Ae . albopictus mosquitoes and further support the hypothesis that this E1-A226V was specifically selected as a result of adaptation of CHIKV to Ae . albopictus mosquitoes . To explain the small fitness advantage associated with the E1-A226V mutation which was observed in transmission experiments , we hypothesize that , similarly to Ae . albopictus , E1-226A and E1-226V viruses colonize different Ae . aegypti organs at different efficiencies . E1-226A appears to colonize midgut cells of Ae aegypti better than E1-226V viruses; however , following dissemination into salivary glands , the E1-226V virus gains an advantage for transmission to vertebrates . The E1-A226V mutation was found to have a slightly negative effect on infectivity , a negligible effect on dissemination , but a slight positive effect on transmissibility of CHIKV by Ae . aegypti in the competition experiment . We suggest that these small ( as compared with Ae . albopictus ) differences associated with the E1-A226V mutation would not be sufficient to have a significant effect on the evolution of CHIKV transmitted by Ae . aegypti and would not result in accumulation of this mutation in the regions where Ae . aegypti serves as a primary vector for CHIKV . This may explain the lack of emergence of the E1-226V genotype in previous outbreaks and the predominance of E1-226A viruses during the 2006 CHIKV epidemic in India , in which Ae . aegypti is considered to be the main vector species [44] . Adaptation of African strains of CHIKV from forest dwelling mosquitoes species to Ae . aegypti has never been shown to be associated with any particular mutations , therefore we believe that the same negative impact of E1-A226V would be seen in African mosquito vectors which were responsible for transmission of CHIKV strains ancestral to Reunion isolates . Our data does not exclude the possibility that the E1-A226V mutation might have a negative effect on the evolution of CHIKV transmitted by Ae . aegypti . Since our dissemination and transmission studies were performed using blood meal titers that were 1–2 Log10TCID50/ml higher than Log10OID50/ml values we suggest that the negative effect of decreased midgut infectivity of E1-A226V on virus transmissibility would be almost completely missed , simply because , under this condition , almost 100% of mosquitoes could become infected . In general , CHIKV requires significantly higher blood meal titers for infection of Ae . aegypti compared to Ae . albopictus [36 , 37] ( Tables 1 and 2 ) , which suggests that the slight decrease in midgut infectivity of E1-226V viruses would have a more profound effect on the evolution of CHIKV transmitted by Ae . aegypti , compared to the effect of a small advantage in the ability to compete with E1-226A viruses for transmission to suckling mice . Therefore , if the E1-A226V mutation occurred in CHIKV transmitted by Ae . aegypti , it would have a weak negative effect on viral fitness and would most likely not be preferentially selected . Additional experiments are required to evaluate this hypothesis . Available data cannot exclude the possibility that E1-226A viruses may have an unknown beneficial effect on the fitness of CHIKV in vertebrate hosts over E1-226V viruses , and that the minor negative effect of E1-226A observed in transmission experiments by Ae . aegypti can be compensated for by more efficient viral replication in the vertebrate host , leading to an overall more efficient adaptation to the transmission cycle . However , comparison of the different effects of A or V residues at position E1–226 on CHIKV infectivity for , and transmission by Ae . aegypti and Ae . albopictus mosquitoes clearly suggests that polymorphisms at this position may determine the host range of the alphaviruses and may play an important role in adaptation of the viruses to a particular mosquito vector . An interesting observation , which should be studied in more detail , was that adaptation of CHIKV to Ae . albopictus mosquitoes coincided with the acquisition of CHIKV dependence on cholesterol in the target membrane . It has been previously shown that various mutations in the same region of the E1 protein of SFV and Sindbis virus can modulate the cholesterol dependence of these viruses [33 , 45] and that SFV independence from cholesterol coincides with more rapid growth of the virus in Ae . albopictus [34] . Although there is an apparent association , it is currently unknown if cholesterol dependence of alphaviruses is directly responsible for modulation of fitness of alphaviruses in mosquito vectors . A possible explanation for the opposite effects of the cholesterol-dependent phenotype of SFV and CHIKV on fitness in Ae . albopictus may reflect the use of different techniques for mosquito infection . In our study , mosquitoes were orally infected via cholesterol rich blood meals , whereas in the previous study SFV was intrathoracically inoculated into the mosquito [34] . It is also possible that cholesterol-dependent and -independent viruses would replicate differently in different mosquito organs . As such , our data indicate that more efficient colonization of Ae . albopictus midgut cells by cholesterol-dependent LR-ApaI-226V is followed by relatively more rapid growth of cholesterol-independent LR −226A virus in mosquito bodies between 3 and 5 dpi ( Figure 4B ) . Three to 5 dpi coincides with virus escape from the mosquito midgut . Alignment of amino acid sequences that constitute the ij loop of E1 protein from different members of the alphaviruses genus revealed that position E1–226 is not conserved ( [33] and data not shown ) and can vary even between different strains of the same virus . In this regard , it would be reasonable to determine the cholesterol requirement of other clinically important alphaviruses , especially Venezuelan equine encephalitis virus ( VEEV ) and eastern equine encephalitis virus ( EEEV ) , which show significant intra-strain variation at position E1–226 among natural isolates of these viruses , and determine mutations which can modulate their cholesterol dependence . In recent studies by Kolokoltsov et al . [46] , it was suggested that VEEV , a New world alphavirus , might be cholesterol independent , although the use of Vero cells instead of C6/36 cells , and the use of different protocols for cell membrane cholesterol depletion , make it difficult to compare the results of this study with our findings . Also it would be of interest to determine possible relationships between mutations which modulate cholesterol dependence of alphaviruses other than CHIKV and on their infectivity for Ae . aegypti and Ae . albopictus mosquitoes and perhaps other epidemiologically important mosquito vectors . The molecular mechanisms responsible for the association between host range and cholesterol dependence of CHIKV are unknown [47] . It has been proposed that upon exposure to low pH , the E1 protein of cholesterol-dependent viruses senses the target membrane lipid composition and goes through a cholesterol-dependent priming recognition reaction [48] which is not required for cholesterol-independent viruses . It is possible that CHIKV infects Ae . aegypti and Ae . albopictus midgut cells using different endocytic pathways , which targets virus to cellular compartments with different lipid contents in which fusion occurs . Specific lipids such as cholesterol may differentially affect fusion of cholesterol-dependent and cholesterol-independent CHIKV strains in these compartments and therefore define the outcome of infection . Although our observations are suggestive , more comprehensive studies should be completed to determine the exact molecular mechanisms responsible for penetration of E1-226A and E1-226V viruses into Ae . aegypti and Ae . albopictus cells . Although previous laboratory studies have demonstrated susceptibility of Ae . albopictus to CHIKV infection [36 , 37] , our data demonstrate that the E1-A226V mutation promoted infection and accelerated dissemination of CHIKV in Ae . albopictus mosquitoes and conferred a selective advantage over infection of Ae . aegypti . Whilst the mutation did not increase the maximum viral titer attainable in the mosquitoes , the synergistic effects of increased infectivity and faster dissemination of the E1-A226V virus in Ae . albopictus would accelerate virus transmission to a naïve human population which would have contributed to initiating and sustaining the 2005–2006 CHIKV epidemic on Reunion island . That a single amino acid change can act through multiple phenotypic effects to create an epidemic situation has implications for other arthropod-transmitted viruses and the evolution of human infectious diseases [49] .
The viruses and plasmids encoding full-length infectious clones of the LR2006 OPY1 strain CHIK-LR ic ( GenBank accession number EU224268; http://www . ncbi . nlm . nih . gov/Genbank/index . html ) and GFP-expressing full-length clone LR-GFP-226V ( CHIK-LR 5′GFP , GenBank accession number EU224269 ) have been previously described [15 , 35] . The plasmids 37997–226A ( pCHIK-37997ic , GenBank accession number EU224270 ) encoding full-length infectious clones of the West African strain of CHIKV 37997 and a GFP-expressing full-length clone 37997-GFP-226A ( pCHIK-37997–5GFP , GenBank accession number EU224271 ) were derived from previously described plasmids pCHIKic and 5′CHIK EGFP [35] by introducing CHIKV encoding cDNA into a modified pSinRep5 ( Invitrogen ) at positions 8055–9930 . Viruses derived from 37997–226A and 37997-GFP-226A are identical to viruses derived form pCHIKic and 5′CHIK EGFP . To facilitate rapid screening of viruses in mosquitoes , the gene encoding enhance green fluorescent protein ( eGFP ) , that is known not to compromise CHIKV phenotype in mosquitoes [15] , was incorporated into clones as previously described [15] . Plasmids were constructed and propagated using conventional PCR-based cloning methods [50] . The entire PCR-generated regions of all constructs were verified by sequence analysis . The maps , sequences and detailed description of the clones are available from the authors upon request . For studies comparing the relative fitness of the mutant ( E1-226V ) virus and the pre-epidemic genotype ( E1-226A ) , a silent mutation ( 6454C ) was introduced into the CHIK-LR ic , to add an ApaI restriction site into the coding sequence of CHIK-LR ic . The resultant plasmid was designated LR-ApaI-226V . The E1-V226A mutation was introduced into CHIK-LR ic and LR-GFP-226V to generate plasmids designated as LR-226A and LR-GFP-226A , respectively . The mutation E1-A226V was also introduced into plasmids 37997–226A and 37997-GFP-226A . The resulted plasmids were designated 37997–226V and 37997-GFP-226V . All plasmids were purified by centrifugation in CsCl gradients , linearized with NotI and in vitro transcribed from the minimal SP6 promoter using the mMESSAGE mMACHINE kit ( Ambion ) following the manufacturer's instructions . The yield and integrity of synthesized RNA were analyzed by agarose gel electrophoresis in the presence of 0 . 25 μg/ml of ethidium bromide . RNA ( 10 μg ) was transfected into 1x107 BHK-21 cells by electroporation as previously described [15] . Cells were transferred to 25 cm2 tissue culture flasks with 10 ml of Leibovitz L-15 ( L-15 ) medium , and supernatants were collected at 24 and 48 h post-electroporation and stored at −80 °C . In parallel , 1x105 electroporated BHK-21 cells were serially 10-fold diluted and seeded in six-well plates for infectious centers assay as previously described [15] . BHK-21 ( baby hamster kidney ) cells were maintained at 37 °C in L-15 medium supplemented with 10% fetal bovine serum ( FBS ) , 100 U penicillin , and 100 μg/ml streptomycin . C6/36 cells ( Ae . albopictus ) were grown in the same medium at 28 °C . Ae . aegypti ( white-eyed Higgs variant of the Rexville D strain ) and Ae . albopictus ( Galveston strain ) were reared at 27 °C and 80% relative humidity under a 16h light: 8h dark photoperiod , as previously described [35] . Adults were kept in paper cartons supplied with 10% sucrose on cotton balls . To promote egg production females were fed on anaesthetized hamsters once per week . Rexville D strain of Ae . aegypti mosquitoes were originally selected for susceptibility to flavivirus infection [51] . Since there are no known consequences of this original selection with respect to susceptibility to CHIKV , a white eyed variant of the strain that facilitates detection of GFP was used in our experiments . To investigate if the mutation influenced cholesterol dependence of the virus , cholesterol-depleted C6/36 cells were prepared by five passages in L-15 medium containing 10% FBS treated with 2% CAB-O-Sil ( Acros Organics ) for 12 h at room temperature as previously described [52] . CHIKV growth curves were determined by infecting cholesterol-depleted and normal C6/36 cells at a multiplicity of infection ( MOI ) of 0 . 1 and 1 . 0 , respectively , by rocking for 1 h at 25 °C . The cells were washed three times with L-15 medium and 5 . 5 ml of fresh L-15 supplied with 10% of standard or CAB-O-Sil treated FBS was added to the flask . At the indicated times post-infection , 0 . 5 ml of medium was removed and stored at −80 °C until titrated . The volume of medium was then restored by adding 0 . 5 ml of appropriate medium . Viral titers from mosquito samples and from tissue culture supernatant were determined using Vero cells and expressed as tissue culture infectious dose 50 percent endpoint titers ( Log10TCID50 ) as previously described [53] . Additionally , for viral competition experiments , titers of LR-Apa-226V LR-226A viruses were determined using standard plaque assay on Vero cells as previously described [54] . Ae . aegypti and Ae . albopictus were infected in an Arthropod Containment Level 3 insectary as described previously [35 , 55] . To make infectious blood meals for the viruses lacking eGFP , viral stocks derived from electroporated BHK-21 cells were mixed with an equal volume of defibrinated sheep blood and supplemented with 3 mM ATP as a phago-stimulant . To produce infectious blood meals for the eGFP-expressing viruses , the viruses were additionally passed on BHK-21 cells . The cells were infected at a MOI ≈ 1 . 0 with virus derived from electroporation . At 2 dpi , cell culture supernatants were mixed with an equal volume of defibrinated sheep blood and presented to 4- to 5-day-old female mosquitoes that had been starved for 24 h , using a Hemotek membrane feeding system ( Discovery Workshops ) and hamster skin membrane . Mosquitoes were allowed to feed for 45 min , and engorged mosquitoes ( stage ≥3+ [56] ) were sorted and returned to a cage for maintenance . Blood meals and three to four mosquitoes were immediately removed for titration and/or RNA extraction . Depending on the purpose of the experiments , mosquitoes were collected at different days post-infection and either titrated to determine viral titer , dissected for analysis of eGFP expression in the midguts or salivary glands [15] , or used for RNA extraction in competition experiments . To estimate the Oral Infectious Dose 50% values ( OID50 ) , serial 10-fold dilutions of viruses were made in L-15 medium followed by mixing the samples with defibrinated sheep blood . Mosquitoes were dissected at 7 dpi and eGFP expression in infected midguts was analyzed by fluorescence microscopy . A mosquito was considered infected if at least one foci of eGFP-expressing cells was present in the midgut . The experiments were performed twice for each virus . OID50 values and confidence intervals were calculated using PriProbit ( version 1 . 63 ) . To test the hypothesis that the E1-A226V mutation might be associated with a competitive advantage in mosquito vectors , competition assays were designed similar to those described previously in mice [57] , with minor modifications ( Figure 2A ) . Both Ae . aegypti and Ae . albopictus mosquitoes were presented with a blood meal containing 107 plaque-forming units ( pfu ) /ml of LR-Apa-226V and 107 pfu/ml of LR-226A viruses . It had been previously found that for these two viruses the ratio of viral RNAs corresponds to the ratio of viral titers ( data not shown ) . Midguts were collected at 7 dpi and analyzed in pools of eight to ten , and heads were collected at 12 dpi and analyzed in pools of five . RNA was extracted from the tissue pools using TRIzol reagent ( Invitrogen ) followed by additional purification using a Viral RNA mini kit ( QIAGEN ) . RNAs from blood meal samples were extracted using Viral RNA Mini Kit followed by treatment with DNAse ( Ambion ) to destroy any residual plasmid DNA contaminant in the viral samples . RNA was reversed transcribed from random hexamer primers using Superscript III ( Invitrogen ) according to the manufacturer's instructions . cDNA was amplified from 41855ns-F5 ( 5′- ATATCTAGACATGGTGGAC ) and 41855ns-R1 ( 5′-TATCAAAGGAGGCTATGTC ) primers using Taq DNA polymerase ( New England Biolabs ) . PCR products were purified using Zymo clean columns ( Zymo Research ) and were quantified by spectrophotometry . Equal amount of PCR products were digested with ApaI , separated in 2% agarose gels that were stained using ethidium bromide . Thus the LR-Apa-226V and LR-226A viruses could be distinguished by size on an agarose gel ( Figure 2A ) . Gel images were analyzed using TolaLab ( version 2 . 01 ) . Relative fitness of LR-Apa-226V and LR-226A viruses was calculated as a ratio between 226V and 226A bands in the sample , divided by the control ratio of 226V and 226A in the blood meal . Ae . aegypti and Ae . albopictus mosquitoes were presented with a blood meal containing 107 pfu/ml of LR-Apa-226V and 107 pfu/ml of LR-226A viruses . At 13 dpi , ten to 15 mosquitoes were placed in separate paper cartons and starved for 24 h . The next day the mosquitoes in each carton were presented with individual 2- to 3-day-old suckling mouse ( Swiss Webster ) . Feeding continued until 2–3 mosquitoes per carton were fully engorged ( stage ≥3+[56] ) . In a parallel experiment six 2- to 3-day-old suckling mice were subcutaneously infected with 20 μl of mixture containing ≈ 25 pfu of LR-Apa-226V and ≈ 25 pfu of LR-226A viruses . Mice were returned to their cage and sacrificed on day 3 post-exposure . Blood from each individual mouse ( ≈ 50 μl ) was collected and immediately mixed with 450 μl of TRIzol reagent for RNA extraction . The RNA was processed as described above . All animal manipulations were conducted in accordance with federal laws , regulations , and in compliance with National Institutes of Health and University of Texas Medical Branch Institutional Animal Care and Use Committee guidelines and with the Association for Assessment and Accreditation of Laboratory Animal Care standards . | Chikungunya virus ( CHIKV ) is an emerging arbovirus associated with several recent large-scale epidemics of arthritic disease , including one on Reunion island , where there were approximately 266 , 000 cases ( 34% of the total island population ) . CHIKV is transmitted by Aedes species mosquitoes , primarily Ae . aegypti . However , the 2005–2006 CHIKV epidemic on Reunion island was unusual because the vector responsible for transmission between humans was apparently the Asian tiger mosquito , Ae . albopictus . Interestingly , the same epidemic was associated with a strain of CHIKV with a mutation in the envelope protein gene ( E1-A226V ) . In this work we investigated the role of the E1-A226V mutation on the fitness of CHIKV in Ae . aegypti and Ae . albopictus mosquitoes . We found that E1-A226V is directly responsible for CHIKV adaptation to Ae . albopictus mosquitoes , which provides a plausible explanation of how this mutant virus caused an epidemic in a region lacking the typical vector . This research gives a new insight into how a simple genetic change in a human pathogen can increase its host range and therefore its geographic distribution . Ae . albopictus is abundant and widely distributed in urban areas of Europe and the United States of America , and this work suggests that these areas are now vulnerable to CHIKV establishment . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] | [
"viruses",
"infectious",
"diseases",
"virology",
"mus",
"(mouse)",
"insects"
] | 2007 | A Single Mutation in Chikungunya Virus Affects Vector Specificity and Epidemic Potential |
Transferrin Receptor ( TfR1 ) is the cell-surface receptor that regulates iron uptake into cells , a process that is fundamental to life . However , TfR1 also facilitates the cellular entry of multiple mammalian viruses . We use evolutionary and functional analyses of TfR1 in the rodent clade , where two families of viruses bind this receptor , to mechanistically dissect how essential housekeeping genes like TFR1 successfully balance the opposing selective pressures exerted by host and virus . We find that while the sequence of rodent TfR1 is generally conserved , a small set of TfR1 residue positions has evolved rapidly over the speciation of rodents . Remarkably , all of these residues correspond to the two virus binding surfaces of TfR1 . We show that naturally occurring mutations at these positions block virus entry while simultaneously preserving iron-uptake functionalities , both in rodent and human TfR1 . Thus , by constantly replacing the amino acids encoded at just a few residue positions , TFR1 divorces adaptation to ever-changing viruses from preservation of key cellular functions . These dynamics have driven genetic divergence at the TFR1 locus that now enforces species-specific barriers to virus transmission , limiting both the cross-species and zoonotic transmission of these viruses .
Transferrin receptor ( TfR1 ) is the cell-surface receptor for iron-loaded transferrin circulating in the blood [1] . TfR1-transferrin complexes are internalized via clathrin-mediated endocytosis and iron is released in acidic endosomes . Besides transferrin , the other major binding partner of TfR1 is the hereditary hemochromatosis protein ( HFE ) , which negatively regulates iron uptake . In addition to these host-beneficial interactions , three different families of viruses are known to interact with TfR1 to trigger their own cellular entry . TfR1 likely constitutes an attractive target for viruses because it is both ubiquitous and specifically up-regulated in rapidly dividing cells [1] . Because of the tremendous investment that has been made in understanding both TfR1 and the viruses that exploit it , there are rich structural and functional data available . For instance , co-crystal structures have been solved of human TfR1 in complex with both of its cellular iron-transport binding partners [2]–[4] and with the surface glycoprotein of a zoonotic rodent arenavirus , Machupo virus , which uses TfR1 for entry [5] . For this reason , TfR1 provides an ideal opportunity to investigate how cellular housekeeping proteins evolve to combat viruses that are exploiting them while simultaneously preserving critical cellular functions . The entry of viruses into cells is often mediated by specific physical interactions between virus surface proteins and host-encoded cell surface receptors . In the case of the New World arenaviruses , the surface glycoprotein , GP , contacts TfR1 to trigger cellular entry [6] . These viruses infect various rodent species found in the Americas , and each virus has evolved compatibility with the particular TfR1 ortholog encoded by its host species ( Figure 1A ) [7]–[9] . Several of these viruses , including Junin virus , Machupo virus , and Guanarito virus , have acquired the ability to bind human TfR1 and are currently emerging into human populations through zoonotic transmission [10] , [11] . These viruses cause hemorrhagic fevers in humans with case fatality rates of 15–30% , but fortunately , they do not yet spread from human to human efficiently enough to cause large epidemics . Another rodent virus that uses TfR1 for cellular entry is the retrovirus mouse mammary tumor virus ( MMTV ) . The MMTV surface glycoprotein , Env , contacts TfR1 to trigger cellular entry [12] . MMTV infects Muridae rodents specifically of the genus Mus , including Mus musculus , the house mouse ( Figure 1A ) . In contrast to the arenaviruses , MMTV is not known to infect other rodent species or humans . Incompatibility with human TfR1 appears to be the major cellular barrier to zoonosis because MMTV replicates robustly in human cells when receptor-mediated entry is bypassed by transfection of the viral genome directly into cells [13]–[15] . Finally , in carnivores , parvoviruses also bind TfR1 for cellular entry [16] . Canine parvovirus serves as one of the most important models for disease emergence in the wild , as this virus first came into existence in the 1970s when a virus was passed to dogs from another carnivore species [17] . This event centered around viral evolution for compatibility with the dog TfR1 ortholog [18] , [19] . Thus , in all three of the virus families that use TfR1 , existing evidence suggests that the ability to enter cells through the TfR1 ortholog of a particular species is a necessary criterion for infection in the wild , and that viral adaptation is often required to utilize the TfR1 of new species . While infectious disease research has long focused on host antiviral proteins , host proteins that facilitate viral replication are now an exploding area of inquiry [20] . These proteins represent novel targets for the development of antiviral drugs because interruption of the interactions between virions and host proteins like TfR1 are predicted to block viral replication . In nature , evolution has utilized two paradigms for achieving this same goal . In some cases , host genes encoding pathogen entry receptors have accumulated promoter or other mutations that cause reduced or no expression of the receptor protein [21]–[27] . However , TFR1 , given the essential nature of its housekeeping functions , would be unlikely to tolerate hypomorphic mutations . For retroviruses , host genomes are known to employ a second mechanism to block virus entry , one that exploits a unique property of the retroviral lifecycle . Unlike other viruses , retroviruses permanently integrate into the host genome during viral replication . If viral genomes become integrated in the host germline , they can be passed to future generations and inherited in a Mendelian fashion [28] , [29] . In several instances , retroviral surface proteins ( Envs ) expressed from these integrated retroviral copies compete with exogenous viruses for receptor use [30]–[35] . Host genomes are presumably selected to keep these retroviral env open reading frames intact because they offer protection against infection by exogenous viruses that use the same receptor [28] , [29] , [36] . Given the critical role of TfR1 in iron homeostasis , there may be a fitness cost to competitive binding by genome-encoded copies of the retroviral Env . Indeed , there is no evidence for either of these models ( hypomorphic mutations or competitive inhibition ) in the TFR1 literature . How , then , do critical genes like TFR1 respond to virus-driven selective pressure ? Most of what is known about the evolutionary dynamics between host and virus genomes comes from studies of antiviral genes , particularly those encoding viral sensors . Viral sensors ( also referred to as “pattern recognition receptors” or “restriction factors” ) are host proteins like RIG-I and TRIM5α that recognize and destroy viruses that are attempting to replicate inside of host cells [37] , [38] . Because these sensors can be so effective , viruses often encode proteins that antagonize them or their downstream executors [39] , [40] . Host genomes are continually selected to encode sensors that better recognize viruses , and viruses are continually selected to evade or disrupt these sensors [41]–[50] . This ongoing evolutionary struggle is called a molecular “arms race” ( reviewed in [51]–[53] ) . Arms races play out in the protein–protein interactions that exist between host and virus proteins , and they drive endless rounds of “positive selection” for mutations that alter these interactions . This results in the rapid evolution of both proteins ( host and virus ) engaged in the conflict . Indeed , host-encoded viral sensors are often exceptionally genetically divergent between species and diverse within species [41]–[50] , [54]–[58] . As a result , such genes are appreciated as major genetic barriers to host switching by viruses in nature , because unique virus mutations are required to counteract the divergent viral sensors present in each new host species [43] , [59]–[61] . Arms races have not traditionally been documented in important housekeeping genes . Here , we document recurrent positive selection in rodent TFR1 and demonstrate that both the protein sequence and the interaction specificities of this receptor are far from static . Using a small evolutionary dataset consisting of TFR1 gene sequences from only seven rodent species , we identify specific codons in TFR1 that have been repeatedly targeted by positive selection for amino acid replacement . We find that these rapidly evolving positions correlate to the surfaces on TfR1 that mediate interaction with the two rodent viruses that bind this receptor . We demonstrate experimentally that mutations at these specific receptor residues are potent at altering interactions with virions while not altering receptor expression or function . We show that this evolutionary scenario has driven genetic divergence at this receptor locus that now enforces species barriers to viral transmission . We address the implications of these findings for human TfR1 and identify a human SNP that conveys some protection against cellular entry of a zoonotic rodent arenavirus . Our study demonstrates that the influence of viral pathogens on mammalian genomes goes well beyond the shaping of antiviral genes , as we can now appreciate that even the sequence of important housekeeping genes can be shaped by unremitting antagonism by viruses . However , in this case , collateral damage to cellular functions must be carefully controlled as the evolutionary battle with viruses plays out .
We investigated the evolution of TFR1 in rodents , where two different virus families use this receptor for cellular entry . The type of selective pressure that has acted on a gene can be inferred from the pattern of mutations that it has accumulated over time [62] , [63] . The rate at which mammalian genes accumulate amino acid–altering DNA mutations ( dN; nonsynonymous mutations ) is typically far slower than the rate at which they accumulate mutations that leave the amino acid unchanged ( dS; synonymous mutations ) [51] . This is because most amino acid–altering mutations are deleterious . This signature ( dN/dS<<1 ) stands in contrast to the pattern that is observed when genes have experienced multiple rounds of positive selection for protein-altering mutations ( dN/dS>1 ) . However , in host-virus arms race situations , patterns of dN/dS>1 would not be expected throughout the entire length of a gene , but rather specifically in the codons correlating to the interaction interface between host and virus proteins ( reviewed in [51] , [52] ) . We used the codeml program in PAML [64] to analyze dN/dS ratios in codons in an alignment of TFR1 from seven rodent species , five of which are known host species for the New World arenaviruses or MMTV ( Figure 1A ) . We found variable patterns of codon evolution in TFR1 . For instance , in codon model M2a , maximum likelihood estimation indicates that 78% of codons are extremely conserved with dN/dS = 0 . 09 , 19% evolve neutrally with dN/dS = 1 , and 2 . 4% are under positive selection with dN/dS = 4 . 2 . Codon models that allow a subset of codons to evolve under positive selection ( dN/dS>1 ) fit the data significantly better than models where positive selection is not allowed ( p<0 . 001; Table S1 ) . Thus , while much of the protein sequence of TfR1 is extremely conserved , a small percentage of residue positions are rapidly evolving . The crystal structure of the TfR1 ectodomain has been solved [65] . Six codons that correspond to residues in this structure were assigned to the dN/dS>1 site class with a high posterior probability: K205 , L209 , N215 , S296 , T569 , and E575 ( Table S1 ) . While discontinuous on the linear polypeptide ( Figure 1B ) , the residues corresponding to these codons are located on a single ridge trailing down the outer edge of each monomer of the human TfR1 dimer ( red residues in Figure 1C ) . Remarkably , all of these sites map precisely to the two known virus-binding surfaces on TfR1 . Three of these rapidly evolving residue positions ( K205 , L209 , and N215 ) map to the arenavirus binding surface of TfR1 ( gray residues in Figure 1C ) [5] . The other three rapidly evolving residues ( S296 , T569 , and E575 ) fall directly in the surface of TfR1 that binds MMTV ( blue residues in Figure 1C ) [13] . We hypothesized that rodent TFR1 is subject to not just one but two different host-virus arms races . Arms races are predicted to drive positive selection in both the host and virus genes involved , so we next analyzed the gene encoding the arenavirus surface protein , GP , for signatures of positive selection . Because the co-crystal structure has been solved of the Machupo virus surface glycoprotein subunit GP1 in complex with TfR1 , the specific residues on GP1 that interact with TfR1 are known ( blue lines below protein schematic in Figure 2A ) . We analyzed an alignment of gp1 from 13 human and mouse isolates of Machupo virus ( Figure 2B ) . In this alignment , 11 codons bear the signature of dN/dS>1 ( red lines above diagram in Figure 2A and Table S2 ) . Ten of these correspond to surface-exposed residues in the GP1 structure [66] . Strikingly , all 10 are located on the surface of GP1 that faces TfR1 , and none fall on the opposite side of GP1 that faces the virion ( Figure 2C ) . Four of the residues under positive selection directly contact TfR1 , and the rest are located near residues that do ( Figure 2C ) . Using a permutation test , we find that the 16 TfR1-binding residues of GP1 are significantly enriched for sites of positive selection ( p<0 . 005 ) . Like all virus surface proteins , GP1 will have also experienced selection for immune escape , a complication that makes signatures of dN/dS>1 more difficult to interpret in viral genes than in host genes . However , GP1 residues in direct contact with TfR1 are unlikely to successfully mutate for the purpose of immune escape during an active infection . An arms race between rodent TfR1 and arenavirus GP1 is thus supported by the rapid evolution of each partner in this interaction , specifically in residues that are known to mediate contact with the other . In the TFR1 dataset analyzed , only one of the rodent species included is known to harbor MMTV in the wild ( house mouse; Figure 1A ) . It was thus unclear why we detected positive selection in the MMTV binding surface of TfR1 with the rodent dataset that was used . We hypothesized that either the evolutionary signature in the MMTV binding surface of TfR1 was driven by something else , or that MMTV-like viruses once circulated more widely through rodent genera . We reasoned that if the latter hypothesis is true , “fossils” of these extinct viruses might be found in the form of endogenous retroviruses ( ERVs ) integrated into the genomes of their former host species . Indeed , we identified MMTV-like ERVs in the genomes of the brown rat ( Rattus norvegicus ) and the North American deer mouse ( Peromyscus maniculatus ) ( Figure 3 and Figure S1 ) . The full-length ERV identified in the deer mouse genome is particularly interesting because this rodent is in the same family as the arenavirus host species ( Cricetidae; Figure 1A ) . These ERVs reveal that MMTV-like viruses once circulated more widely amongst rodents , supporting the model that rodent TFR1 may have experienced selection imposed by these viruses . Interestingly , MMTV appears to be a virus in retreat , with a shrinking host range . We cannot exclude the possibility that MMTV-like viruses still infect other rodent species and have simply not been identified , but such viruses have not been reported in the literature or in GenBank [67] , and are absent from large metagenomic surveys of rodent feces [68] . These MMTV ERVs are thus reminiscent of the many ERV families found in the human genome , none of which currently circulate in infectious form [28] . Based on these findings , TfR1 may have experienced high levels of sequence divergence on the MMTV-binding surface due to selection for mutations that blocked entry by these MMTV-like viruses . Consistent with this , we find that TfR1 orthologs from three different Cricetidae species are highly recalcitrant to entry by MMTV ( Figure 4 ) , even though this rodent family appears to once have harbored a similar virus . In an arms race between TfR1 and MMTV , the MMTV Env should also be evolving in response to the evolution of TfR1 . Compared to Machupo virus GP1 , far less is known about the amino acids in MMTV Env that bind to TfR1 , as there is no co-crystal structure of Env in complex with TfR1 . However , a five amino acid receptor binding motif in MMTV Env has been identified [69] . We find that this motif has a distinct protein sequence depending on the particular rodent host species from which each virus was isolated ( Figure S2 ) , consistent with viruses having uniquely evolved compatibility with each host TfR1 ( before they potentially went extinct ) . An incomplete understanding of receptor binding determinants in MMTV Env , and the fact that most of these viruses now exist as endogenous copies , make it difficult to draw specific conclusions about the evolution of MMTV Env . Nonetheless , an arms race between TfR1 and MMTV is supported by the rapid evolution of residues on the MMTV-interaction surface of TfR1 , the discovery that MMTV-like viruses once infected rodents more broadly providing a model for what drove this selection , and the observation that several Cricetidae TfR1 in their current form do not support MMTV entry , suggesting that they could have been selected for this property . To test this MMTV resistance hypothesis further , we simulated the evolution of an MMTV-resistant receptor by mutating only the residue positions under positive selection in the MMTV binding surface ( Figure 5A ) . We mutated the TfR1 of house mouse , the MMTV host , so that these three positions now encode the amino acids found in the TfR1 of the vesper mouse , which is not susceptible to MMTV . MDCK ( dog ) cells were transduced to stably express the mutant or wild-type TfR1 protein . These cells were chosen because dog TfR1 does not support entry by arenaviruses [9] or MMTV [13] . An extracellular FLAG tag was added to each receptor so that cell surface expression could be monitored on live cells by flow cytometry . We then measured the cellular entry of GFP-encoding retroviral vectors expressing the MMTV Env on their surface ( MMTV pseudoviruses ) . Indeed , the three mutations in house mouse TfR1 almost completely abolished the entry of MMTV into cells ( Welch t-test , p<0 . 0001 , one-tailed; Figure 5B ) without significantly altering receptor cell surface expression ( Figure 5C ) . None of the sites of positive selection that we identified are found near the dimerization domain of TfR1 , the region known to be most important for interaction with iron-transport binding partners ( Figure 6A , B ) [2]–[4] , [70] . We confirmed that these mutations indeed do not alter transferrin binding ( Figure 6C , D ) . Thus , amino acid substitutions at these sites in TfR1 can block virus entry without deleterious consequences to surface expression or receptor function , providing a clear hypothesis for why they might have a strong selectable advantage in MMTV-infected rodent populations . If positively selected residues are key modulators of virus compatibility , we reasoned that mutations at these sites should also render MMTV-resistant TfR1s susceptible to MMTV entry . Because species divergence can lead to subtle structural differences in receptors , creating a gain-of-function phenotype with just three amino acid changes should be substantially more difficult than creating a loss of function phenotype in a receptor where virus-binding is currently intact . Nonetheless , mutating the three positively selected residues in the MMTV binding surface of zygodont TfR1 to match the corresponding residues found in TfR1 of house mouse ( the MMTV host ) led to a significant increase in MMTV entry ( Welch t-test , p = 0 . 008 , one-tailed; Figure 5D ) without enhancing cell-surface expression ( Figure 5E ) , transferrin binding ( Figure 6C , D ) , or entry of three arenaviruses ( Figure 5F ) . Thus , we have shown that swapping amino acids encoded at positively selected sites can swap virus-susceptibility phenotypes of TfR1 in both a gain-of-function and loss-of-function manner . Mutations at just three residue positions acutely regulate virus entry while preserving receptor expression and transferrin binding for the host . Every round of positive selection of the rodent TFR1 gene began with a random mutation that arose in a single rodent individual . If this mutation offered protection against virus entry while not otherwise causing major fitness defects related to iron homeostasis , it would have been favored by natural selection and would have become more common or even fixed in the population where it arose . Because the New World arenaviruses are currently emerging into human populations , they are now beginning to exert selective pressure on the human population as well . For instance , there have been approximately 30 , 000 cases of Argentine hemorrhagic fever caused by the Junin virus since the 1950s , with a case fatality rate of 20% [11] . The geographic region at risk for this disease is expanding into north-central Argentina , and currently includes an area populated by around 5 million people [11] . Individuals with genotypes that make them less susceptible to infection or severe illness are expected to survive with bias over other individuals . This selection would intensify as the frequency or severity of the disease increases . In such cases , natural selection would be expected to act at any genetic locus where functionally distinct alleles exist within the human population . We wished to investigate whether TFR1 may be one such locus . TfR1 interacts with arenaviruses and MMTV through distinct interaction surfaces ( Figure 1C ) . TfR1 is 760 amino acids long , but a small stretch of nine residues from 204 to 212 is the major determinant of species-specificity for arenavirus entry ( colored yellow in Figure 7A ) . These residues span two beta strands and the intervening loop ( βII-1–βII-2 ) . Two of the sites of positive selection ( residues 205 and 209 ) fall in this stretch of nine residues , and the third ( residue 215 ) falls three amino acids away ( colored red in Figure 7A ) . As we demonstrated for the sites under positive selection in the MMTV binding surface , the introduction of amino acids from different rodent species at positions in this stretch has been previously shown to alter patterns of virus compatibility [7] , [8] . Additionally , substitution of rodent-encoded amino acids at these residues can convert human TfR1 into an entry receptor for currently non-zoonotic rodent arenaviruses [5] , [8] . By querying SNP databases , we identified a human SNP located in this structural feature , L212V ( colored blue in Figure 7A ) . Because of the localization of this SNP near the residues under positive selection , we hypothesized that the L212V human polymorphism might affect arenavirus entry . To test this , we again focused on Machupo virus . We constructed stable cell lines that express either human 212L or 212V TfR1 . In the context of MDCK cells , dog TfR1 does not allow entry by Machupo virus , so the expression of either human allele allows more virus entry than is observed in MDCK cells alone ( Figure 7B ) . However , the minor TfR1 212V variant supports about half the level of entry as seen with TfR1 212L ( Figure 7B ) . Valine at position 212 may lead to a modest decrease in binding affinity with GP1 due to loss of a hydrophobic contact , based on the observation that two residues of Machupo GP1 ( Phe226 and Pro223 ) are in van der Waals contact with TfR1 Leu212 [5] . We next stably expressed the human 212V and 212L TFR1 alleles in human cell lines that are themselves homozygous for 212L: HEK293 ( kidney ) and HEL299 ( lung ) . Lung cells are especially relevant since arenaviruses are transmitted to humans through respiratory inhalation . In both cases , expression of the minor 212V allele was again protective against virus entry compared to the wild-type allele ( Figure 7C , D ) . Thus , we have identified a SNP ( L212V ) that conveys some protection against arenavirus entry , at least in vitro . The L212V SNP has only been reported in Asian populations ( Chinese and Japanese ) , while TfR1-utilizing arenaviruses have only been found in the Americas . We sequenced TFR1 from 18 indigenous Central and South American individuals , but identified no instances of this polymorphism . Like all SNPs , this SNP arose randomly and may have no fitness advantage or disadvantage in the Asian populations where it is found , since TfR1-utilizing arenaviruses are not found in that part of the world . Nonetheless , this SNP could begin to experience selection if the rodent populations that carry these viruses were introduced into Asia , if these arenaviruses ever evolved to spread efficiently from human to human , or in the event of an intentional release of these viruses [71] . The data shown in Figure 7C , D indicate that protective TFR1 alleles can act in a semidominant fashion with regards to virus entry , because the human cells used in these experiments also express wild-type TfR1 . We speculate that this occurs either because mutant and wild-type TfR1 proteins are forming heterodimers with one another , or because expression of a second allele that is functional for iron-uptake results in lower levels of wild-type TfR1 ( TFR1 expression levels are tightly regulated for the purpose of maintaining iron homeostasis [10] ) . Either model would also be relevant in heterozygous individuals , suggesting that selection could act on SNPs conveying protection against viral entry even when they are rare and found predominantly in heterozygotes .
In this study we show that the protein sequence and interaction specificities of rodent TfR1 have been dynamic over time , shaped by selective pressures imposed by viruses . These dynamics have played out through mutations accumulated at just a small number of residue sites , where mutations decrease virus entry without measurably affecting receptor expression or iron-transport functions . TFR1 represents the first case , to our knowledge , where the evolution of a single host gene is driven by two host-virus arms races at once . In the case of the MMTV binding surface , this has played out through three residue positions coordinated in three-dimensional space . In the arenavirus binding surface , the target of selection has been a small surface-exposed structural feature , in which we were able to detect positive selection of three of the residues . Outside of rodents , TfR1 is used by a third family of viruses , the parvoviruses , and carnivore TFR1 is also under positive selection [72] . TFR1 evolution has thus been shaped by viruses in two separate species groups ( rodents and carnivores ) and by every viral pathogen known to use this receptor . These findings now explain how TFR1 became divergent enough to create species-specific interactions with all three of these virus families . If even a few residue positions can evolve to block virus entry without collateral damage to cellular function , host-virus arms race dynamics can unfold even in genes encoding highly conserved and essential housekeeping proteins . This evolution of TFR1 can be put into contrast with other types of pathogen-driven positive selection of host genes . The human CCR5 gene encodes a co-receptor for HIV cellular entry . Some humans encode a variant allele of CCR5 , CCR5Δ32 , where a 32 base pair deletion gives rise to a defective receptor that is not expressed on the cell surface [73] . Individuals homozygous for this allele are almost completely resistant to HIV infection , and even heterozygous genotypes afford some protection due to reduced expression of wild-type CCR5 . Like the model proposed herein for TFR1 L212V , CCR5Δ32 pre-dates HIV and may or may not have had any functional significance before the HIV pandemic . Nonetheless , it has become highly relevant in a world with HIV/AIDS . Like HIV , most simian immunodeficiency virus ( SIV ) strains also use CCR5 as a co-receptor . In a fascinating case of convergent evolution , some sooty mangabeys and red-capped mangabeys also encode null or defective alleles of CCR5 [21] , [23] . Similarly , the DARC gene encodes a chemokine receptor that is used as an entry receptor by some malaria-causing Plasmodium species . A cis-regulatory polymorphism that silences DARC expression in erythrocytes has arisen independently in human populations from different parts of the world and is highly protective against Plasmodium vivax and Plasmodium knowlesi infection [24] , [25] . Similar mutations have arisen in the cis-regulatory region of DARC in African baboons , and these are associated with resistance to a malaria-like parasite common in baboon populations [26] . In all of these cases , it has been speculated that selective pressure exerted by pathogens has driven these hypomorphic receptor alleles to high frequency in the affected human and nonhuman primate populations . These CCR5 and DARC examples represent a more common mode of pathogen-driven positive selection ( not recurrent ) than the one demonstrated for TFR1 , and there are several important differences . When receptor genes experience hypomorphic mutations , the predominant evolutionary strategy available to viruses will be to use a new receptor altogether . Indeed , the SIV strains that infect sooty and red-capped mangabeys ( SIVsmm and SIVrcm ) have both evolved to use alternate co-receptors [21] , [23] . A few CCR5Δ32 homozygous humans have also been reported to be infected with HIV , again through mutations that allow the virus to use an alternate co-receptor ( CXCR4 in this case ) . Hypomorphic mutations in receptors are not expected to be “serially replaced” due to arms race dynamics . Rather , viral evolution to use a new receptor ends the arms race with the original receptor gene and starts a new one with the new receptor gene . The CCR5 and DARC examples also involve evolutionary time scales millions of years shorter than what has been demonstrated in the current study; because these hypomorphic alleles are circulating in populations of individuals and are not shared between species , they have arisen relatively recently . Also , because these mutations simply reduce cellular expression of the encoded receptors , they presumably have some negative fitness effect on the host . The TfR1 example that we provide here is unique because solutions to viral entry have been found that appear to lack collateral damage to transferrin binding , and presumably to other host functions as well . Because of this , these mutations become common or fixed in populations where they occur , and are serially replaced as viruses continue to evolve and as rodents continue to speciate . There is reason to believe that host-virus arms races are also shaping the protein sequence of other virus entry receptors in the manner described here . There are several other examples where significant sequence and functional divergence exist both on the side of a virus and its host entry receptor . For instance , certain strains of murine leukemia virus ( MLV ) use the rodent XPR1 receptor for cellular entry [74] . There are several functionally distinct variants of the XPR1 gene encoded by rodents of the genus Mus , each with its own pattern of virus susceptibilities . The viruses that use this receptor are also highly variable in the receptor-binding portion of their surface protein , Env . High levels of sequence divergence and disparate interaction specificities have also been observed between the entry receptor TVB encoded by birds and the avian leukosis virus ( ALV ) strains that use this receptor [75] . In neither of these cases is the housekeeping function or structure of the receptor known , so the pleiotropic consequences of pathogen-driven selection remain to be explored . However , both of these viruses can evolve to use new allelic forms of their receptor encoded by new hosts , suggesting that the receptors are important determinants of host range . High levels of sequence divergence , along with polymorphic and species-specific interactions between receptors and viruses , should be the hallmark for this type of evolution . These patterns have also been observed in other pairs of receptors and viruses [72] , [76]–[80] , suggesting that arms races might shape many receptors and potentially other types of housekeeping proteins exploited by viruses as well [81] , [82] . Traditionally , TfR1 has been viewed as a housekeeping protein with an immensely important and conserved role in the cell . This study provides a much richer understanding of the multiple dynamic roles that this receptor is balancing in nature .
Rodent TFR1 and Machupo gp1 sequences were analyzed for positive selection . Database accession numbers for sequences used are listed in Tables S1 and S2 . Sequences were aligned in Clustal [83] , with minor adjustments made by hand ( these two alignments contain few or no indels , respectively ) . jModeltest v2 . 1 . 1 [84] was used to select the best-fit model of nucleotide substitution , which was HKY+G in both cases . Phylogenetic trees for each sequence set were built by the maximum likelihood method implemented in MEGA5 [85] . The TFR1 gene tree matches the species tree of these rodents [86] . Because the Machupo gp1 sequences represent viral isolates from the same population , GARD [87] was run on the gp1 alignment to confirm the lack of phylogenetic breakpoints indicative of recombination . For both datasets , maximum likelihood analysis of dN/dS was then performed with codeml in the PAML 4 . 1 [64] software package . To detect selection , multiple alignments were fit to the NSsites models M1a ( neutral model , codon values of dN/dS are fit into two site classes , one with value between 0 and 1 , and one fixed at dN/dS = 1 ) , M2a ( positive selection model , similar to M1a but with an extra codon class of dN/dS>1 allowed ) , M7 ( neutral model , codon values of dN/dS fit to a beta distribution , dN/dS>1 disallowed ) , M8a ( neutral model , similar to M7 except with a fixed codon class at dN/dS = 1 ) , and M8 ( positive selection model , similar to M7 but with an extra class of dN/dS>1 allowed ) . Model fitting was performed with multiple seed values for dN/dS ( ω ) and assuming either the f61 or f3x4 model of codon frequencies [88] . Likelihood ratio tests were performed to assess whether permitting some codons to evolve under positive selection gives a significantly better fit to the data than models where positive selection is not allowed . The results obtained were shown to be robust to changes in the codon frequency model used , and the seed value for dN/dS ( Tables S1 and S2 ) . Posterior probabilities of codons under positive selection in M8 were then inferred using the Naive Empirical Bayes ( NEB ) algorithm . Coordinates for molecular structures were obtained from the RSCB protein database ( http://www . pdb . org/ ) and rendered using PyMOL ( http://www . pymol . org ) . Full-length MMTV sequences were obtained on GenBank ( AF228552 , D16249 , AF033807 , AF228551 ) . These sequences were used to BLAT [89] the current assemblies of the M . musculus ( mm9 ) [90] and R . norvegicus ( rn4 ) [91] genomes on the UCSC genome browser [92] , recovering the indicated ERVs in these genomes . The nr/nt database for rodents ( taxid:9989 ) at NCBI was searched for similar sequences in other species using the discontiguous megablast search algorithm with full-length MMTV as a query , and using the tBLASTx algorithm with MMTV pol as a query . Both of these approaches identified the Peromyscus maniculatus ERV buried in the sequence of GenBank record EU204642 ( a BAC clone containing the deer mouse beta-globin gene cluster ) . A relatively young age of this ERV can be inferred from the fact that one open reading frame ( pol ) is still uninterrupted , and from the observation that the 5′ and 3′ LTRs differ at only 1 out of 917 positions . The giraffe , bison , and musk ox sequences are from [93] . Exogenous and endogenous beta-retrovirus genome sequences were aligned with MUSCLE [94] as implemented in MEGA5 [85] . jModeltest v2 . 1 . 1 [84] was used to select GTR+I+G as the best-fit model of nucleotide substitution . Phylogenetic trees were built by the maximum likelihood method implemented in MEGA5 . Positions in which one or more sequences contained a gap were excluded during tree building . One thousand bootstrap replicates were performed and results are presented as percentage of replicates that supported each node . The L212V SNP in human TFR1 ( rs41301381 ) was identified in data deposited by the 1000 Genomes Project ( http://browser . 1000genomes . org ) . As of Release 12 , L212V had been found as a heterozygous SNP in 11 individuals , with no homozygous carriers identified . Three of these individuals were Han Chinese from the South ( CHS population ) , six were Han Chinese from Beijing ( CHB population ) , and two were Japanese individuals ( JPT population ) . In total , 11 out of 286 Asian individuals surveyed were heterozygous at this position , yielding a genotypic frequency of 0 . 038 in Asia . This SNP has not been included in the HapMap Genotyping Project ( as of Release 28 ) . Human embryonic kidney 293T cells ( ATCC CRL-11268 ) , HEK293 cells ( ATCC CRL-1573 ) , human embryonic lung HEL299 cells ( ATCC CCL-137 ) , and canine kidney MDCK . 2 cells ( ATCC CRL-2936 ) were all maintained in Dulbecco modified Eagle's medium ( Cellgro ) supplemented with 10% fetal bovine serum ( Gibco ) , 100 units ml−1 penicillin , 100 µg ml−1 streptomycin , and 2 mM L-glutamine ( Cellgro ) . Human , Mus musculus , Calomys musculinus , Calomys callosus , and Zygodontomys brevicauda TFR1 with an encoded C-terminal FLAG tag were moved from pcDNA3 . 1 ( + ) vectors ( described previously [7] ) into the Gateway entry vector pCR8 using the pCR8/GW/TOPO TA Cloning Kit ( Invitrogen ) . The following primers were used to amplify TfR1 for TA cloning: 5′-TTAATACGACTCACTATAGGG-3′ and 5′-TAGAAGGCACAGTCGAGGC-3′ . Gateway LR recombination ( Invitrogen ) was performed to transfer TFR1 genes from pCR8 into the entry site in a Gateway-converted LPCX retroviral vector . Site-directed mutagenesis of the human , M . musculus , and Z . brevicauda TFR1 orthologs was performed using QuikChange Site-Directed Mutagenesis kit ( Stratagene ) . Plasmids encoding Machupo , Junin , and Guanarito GP have been described previously [6] . An MMTV Env-encoding plasmid ( pQ61 ) was kindly provided by Dr . Susan Ross ( via Dr . Jackie Dudley ) . The above described LPCX:TFR1 retroviral vectors were packaged in 293T cells by co-transfecting them along with the NB-MLV packaging plasmid pCS2-mGP [95] and pC-VSV-G using Fugene ( Roche ) . Supernatants were collected and used to infect MDCK . 2 ( dog ) cells . After 24 h , media containing 3 . 5 µg ml−1 puromycin was added to select for transduced cells ( 1 . 0 µg ml−1 puromycin was added when creating the HEK293 and HEL299 stable cell lines ) . These receptors have a C-terminal FLAG tag that is extracellular when the receptor is at the cell surface [8] . Expression of TfR1 proteins was detected in live cells by flow cytometry using an anti-FLAG antibody conjugated with Allophycocyanin ( Abcam , catalog ab72569 ) . Stable cell lines expressing human 212L and 212V TFR1 alleles were made in MDCK , HEK293 , and HEL299 cells as described above . Arenavirus GP or MMTV Env pseudotyped MLV recombinant retroviruses were packaged in 293T cells . Fugene ( Roche ) was used to co-transfect the GFP-encoding transfer vector pQCXIX ( BD Biosciences ) along with plasmids encoding MLV Gag-Pol and one of the viral surface glycoproteins Machupo GP , Junin GP , Guanarito GP , or MMTV Env . After 48 h , supernatants containing viruses were harvested , filtered , and frozen at −80°C . For entry assays , cell lines stably expressing various TfR1 orthologs or human alleles were plated at a concentration of 1 . 0×105 cells per well in a 24-well plate and , after 24 h , infected with pseudotyped virus along with 5 µg ml−1 polybrene . The plates were spinoculated with centrifugation at 350g for 1 . 25 h at 30°C . After 2 h of incubation at 37°C , cells were washed once with PBS and the media was replaced . Two days postinfection , cells were analyzed by flow cytometry . Cells were first gated for live cells and then , using an anti-FLAG antibody conjugated with Allophycocyanin ( APC; Abcam , catalog ab72569 ) , further gated such that all samples were narrowed to the same log decade of receptor expression ( capturing the majority of cells but excluding outliers ) . Where TfR1 expression levels are reported , this is the mean fluorescent intensity within this gated population ( 10 , 000 cells ) . These same 10 , 000 cells were scored for expression of GFP ( viral entry ) . Analysis of flow cytometry data was performed using FlowJo 8 . 8 . 6 ( TreeStar Inc , Ashland , OR ) . MDCK . 2 stable cell lines expressing FLAG-tagged TfR1 orthologs were trypsinized and aliquoted in triplicate at a concentration of 2 . 5×105 cells/tube . The cells were washed with DPBS with 1% ovalbumin ( Sigma ) . The cells were then resuspended in 200 µL of DPBS with 1% ovalbumin containing 1∶500 dilution of FITC-conjugated Mouse transferrin ( 2 . 0 mg/mL stock concentration; Jackson ImmunoResearch , 015-090-050 ) and incubated at 37°C for 60 min . Anti-DDDDK ( FLAG ) tag antibody conjugated with Allophycocyanin ( 0 . 1 mg/mL stock concentration; Abcam , catalog ab72569 ) was added to the cells at a 1∶100 dilution and incubated on ice for 20 min . The cells were then washed twice , resuspended in DPBS with 1% ovalbumin , and then analyzed by flow cytometry . Cells were first gated for live cells and then further gated such that all samples were narrowed to the same log decade of receptor expression ( capturing the majority of cells but excluding outliers ) . Where TfR1 expression levels are reported , this is the mean fluorescent intensity within this gated population ( 10 , 000 cells ) . These same 10 , 000 cells were simultaneously analyzed for transferrin binding with FITC . Analysis of flow cytometry data was performed using FlowJo 8 . 8 . 6 ( TreeStar Inc . , Ashland , OR ) . | Genetic differences between mammalian species dictate the patterns of viral infection observed in nature . They also define how viruses must evolve in order to infect new mammalian hosts , giving rise to new and sometimes pandemic diseases . Because viruses must enter cells before they can replicate , new diseases often emerge when existing viruses evolve the ability to bind to the cell-surface receptor of a new species . At the same time , host cell receptors also evolve to counteract virus attacks . This back-and-forth evolution between virus and host can lead to an arms race that shapes the sequences of the proteins involved . In wild rodent populations , the retrovirus MMTV and New World arenaviruses both exploit Transferrin Receptor 1 ( TfR1 ) to enter the cells of their hosts . Here we show that the physical interactions between these viruses and TfR1 have triggered evolutionary arms race dynamics that have directly modified the sequence of TfR1 and at least one of the viruses involved . Computational evolutionary analysis allowed us to identify specific residues in TfR1 that define patterns of viral infection in nature . The approach presented here can theoretically be applied to the study of any virus , through analysis of host genes known to be key to controlling viral infection . As such , this approach can expand our understanding of how viruses emerge from wildlife reservoirs , and how they drive the evolution of host genes . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"virology",
"emerging",
"viral",
"diseases",
"genetics",
"molecular",
"genetics",
"host-pathogen",
"interaction",
"biology",
"microbiology",
"evolutionary",
"biology",
"evolutionary",
"genetics"
] | 2013 | Dual Host-Virus Arms Races Shape an Essential Housekeeping Protein |
Investigators have linked rare copy number variation ( CNVs ) to neuropsychiatric diseases , such as schizophrenia . One hypothesis is that CNV events cause disease by affecting genes with specific brain functions . Under these circumstances , we expect that CNV events in cases should impact brain-function genes more frequently than those events in controls . Previous publications have applied “pathway” analyses to genes within neuropsychiatric case CNVs to show enrichment for brain-functions . While such analyses have been suggestive , they often have not rigorously compared the rates of CNVs impacting genes with brain function in cases to controls , and therefore do not address important confounders such as the large size of brain genes and overall differences in rates and sizes of CNVs . To demonstrate the potential impact of confounders , we genotyped rare CNV events in 2 , 415 unaffected controls with Affymetrix 6 . 0; we then applied standard pathway analyses using four sets of brain-function genes and observed an apparently highly significant enrichment for each set . The enrichment is simply driven by the large size of brain-function genes . Instead , we propose a case-control statistical test , cnv-enrichment-test , to compare the rate of CNVs impacting specific gene sets in cases versus controls . With simulations , we demonstrate that cnv-enrichment-test is robust to case-control differences in CNV size , CNV rate , and systematic differences in gene size . Finally , we apply cnv-enrichment-test to rare CNV events published by the International Schizophrenia Consortium ( ISC ) . This approach reveals nominal evidence of case-association in neuronal-activity and the learning gene sets , but not the other two examined gene sets . The neuronal-activity genes have been associated in a separate set of schizophrenia cases and controls; however , testing in independent samples is necessary to definitively confirm this association . Our method is implemented in the PLINK software package .
Multiple recent studies have demonstrated a convincing and statistically significant excess of rare CNVs in individuals affected by schizophrenia compared to unaffected individuals [1]–[4] . Similar observations have now been made separately in autism [5]–[7] and bipolar disorder [8] . However , it is typically not readily evident which individual CNV events are pathogenic since ( 1 ) many rare events are seen in the general population and the excess in cases is relatively modest and ( 2 ) individual events are too rare to demonstrate definitive association in realistically sized patient collections . Hence , it is challenging to translate these rare CNV events into a clear understanding of disease pathology . To identify candidate genes for follow-up , investigators have employed statistical tests of gene set enrichment , originally developed as an effective approach to interpret gene expression data [9] . Practically , these analyses identify functional gene sets or ‘pathways’ that are over-represented among those genes affected by case CNVs compared to unaffected genes [1] , [8] , [10] , [11] , often relying on online resources such as Panther [12] , Ingenuity Pathway Analysis ( Ingenuity Systems , www . ingenuity . com ) , and Gene Ontology [13] . For example , gene set enrichment analyses by Walsh et al . suggested that rare CNVs in schizophrenia preferentially disrupt those genes with neuro-developmental functions [1]; Zhang et al . similarly reported that rare CNVs in bipolar disorder preferentially overlap genes involved in behavior and learning [8] . More recently Glessner et al . reported that genes affected by rare and common CNVs in autism are also involved in brain function [11] . While these initial results are highly promising , the gene set enrichment statistical framework as applied to copy number variation is critically limited and potentially confounded . The key analytical question in this setting is whether an event that impacts a set of genes or a pathway , increases disease risk compared to events that do not impact that pathway . Under the hypothesis that events affecting a specific brain-function pathway are pathogenic , the rate of those events affecting the brain-function pathway should indeed be greater in cases than in controls – ideally fully explaining the observed genome-wide differences in case and control event rates . An alternative possibility is that the increased rate and size of CNVs in cases represents a mutational syndrome or genomic instability , and that they are not in themselves pathogenic . Under that possibility , case events should not preferentially impact any particular gene set; however , differences is size and rate might be observed . The commonly used gene set enrichment analytical approach used to address this question falls short on two accounts: ( 1 ) they do not rigorously compare case event rates to control event rates and ( 2 ) since they examine affected genes rather than events , they do not accurately account for the fact that multiple genes might be contributed from a single event or that single genes may be affected by multiple events . Here we propose a straight forward statistical test to explicitly compare the rate of CNVs impacting a specific gene set in cases to the rate in controls that carefully accounts for background differences in CNV rate and size . A possible consequence of not rigorously comparing event rates in cases to controls is that sets consisting of genes that are more frequently affected by CNVs might spuriously appear to be highly enriched in cases , but also will be highly enriched in controls . Examples of such genes include large genes spanning a massive portion of the genome or those whose functions are highly redundant or non-essential . This is a particularly important issue for neuropsychiatric disease considering the reportedly large size of genes with brain function . Multiple studies of CNVs in the general population have reported enrichment for neuro-physiological genes [14] , [15] – suggesting that brain-function genes may be susceptible to CNV events in general , possibly due to their large size or other factors . In fact , published events in neuropsychiatric disease studies often implicate large genes ( see Table S1 ) . Others have already noted that gene size itself can bias pathway enrichment analyses in other contexts , such as annotating non-coding elements for function [16] , [17] . In particular , Taher et al noted that randomly selected positions in the genome are enriched for proximity to genes involved in “development” , “cell-adhesion” , and “nervous system development” [17] . Some of the published disease studies attempted to address this issue indirectly by applying similar analyses to control events and note the lack of statistical evidence of enrichment for brain function gene sets [1] , [8]; however , control events are typically fewer and smaller , implicating many fewer genes , and therefore simply comparing the statistical significance of gene set enrichment results in cases and controls is not adequate . One possible consequence of examining genes rather than the events they occur in is that individual large events contribute many genes and may skew the analysis much more than smaller events , and cause spurious findings . This is of particular concern since genes with common function can often cluster together on the genome and a single event in one individual affecting a cluster of related genes can naively appear to implicate an entire pathway [18] , [19] . One interesting example is the reported enrichment of psychological disorder genes in the Zhang et al data set ( see Table S1 ) ; 11 of the 16 deleted psychological disorder genes are in the 22q11 . 21 region observed in two individuals in the data set [8] . These genes are possibly annotated as psychological disorder genes since rare deletions in 22q11 . 21 have long been observed among schizophrenia cases [2] . Removal of the two individuals with 22q11 . 21 events eliminates any enrichment for the psychological disorder gene set – suggesting that there is little evidence that this particular set is necessarily relevant to disease outside of the 22q11 . 21 region . Of course at least one gene in this region is pathogenic , but it is unlikely that >10 in this region are and that in aggregate define a key pathogenic set . A second possible consequence of examining genes and not the events they occur in is that genes affected in both cases and controls , but at different rates , are not properly accounted for . For instance , a critical gene affected by many pathogenic events contributes equally to a gene set enrichment analyses as a gene sporadically affected by a single event . One interesting example is the NRXN1 gene , a large gene that plays an important role in synaptic development [20] . Since CNV events affecting NRXN1 have been observed in both schizophrenic cases and controls , they would contribute equally to a pathway analysis of case events as they would to one of control events . However , the rate of functional events observed in cases is significantly more than in controls; pathway-based approaches could be bolstered if methods explicitly take into account these differences between cases and controls event rates for genes of interest . Here , we describe a case-control statistical test , cnv-enrichment-test , to explicitly compare the rate of CNVs impacting specific genes sets in cases to controls . We show how cnv-enrichment-test is robust to even extreme biases in gene size and case-control differences in CNV rate and size . We also demonstrate how standard gene set enrichment approaches is often confounded under realistic scenarios , by gene size and other gene structural features; we demonstrate these confounders in a set of 2 , 415 controls genotyped for rare single-event deletions . We finally apply the cnv-enrichment-test to examine genes with brain function within a large dataset of CNVs identified in schizophrenia cases and controls published by the International Schizophrenia Consortium ( ISC ) [2] and demonstrate nominal evidence of association for previously described gene sets .
Set enrichment is the standard approach to test whether genes impacted by CNVs in cases affect specific pathways . Specifically , the overlap between the set of genes affected by CNVs is compared to the set of genes with a particular function . Genes affected by a CNV might be defined as disrupted genes or overlapped genes . Disrupted genes are those genes that have a CNV boundary that falls within the boundaries of its transcript [1] . Overlapped genes are a superset of those genes whose transcripts are either disrupted by a CNV or are fully contained by a CNV [8] . Since genes rarely overlap each other , a single CNV event might contribute up to two disrupted genes but many overlapped genes . Both have been previously examined in the literature . Unless otherwise specified , this study emphasizes overlapped genes . After identifying the genes affected ( overlapped or disrupted ) by a CNV , we then identify genes with a specific process or within a specific pathway . We apply a two-tailed Fisher's test to assess whether the number of affected pathway genes is statistically significantly different than might be affected by chance . The critical assumption in gene set-based analyses is that there is a single independent observation per gene , not connected to the gene's size or structural features . As an alternative , we propose a simple case-control strategy to test gene sets or pathways for association to disease: the “cnv-enrichment-test” . This strategy is consistent with the case-control association framework used in CNV and SNP disease association studies [21] , [22] . A direct case-control comparison avoids any ascertainment bias that might be the consequence of structural features of genes within a set , since the same biases will apply equally to both cases and controls . We are careful to control for case-control differences in CNV rate and size , since those differences can artificially induce a pathway association . For example , if the rare CNV rate in cases is more frequent or larger than in controls , then on average all genes will be impacted more often in cases , and any arbitrary gene set might appear to be affected more commonly in cases than in controls . Also , if rare CNVs are smaller but more frequent in cases than in controls , then sets of larger genes might appear to be impacted more often in cases than in controls . To assess whether CNV events specifically overlapping genes in the pathway of interest are enriched in cases compared to controls , we propose the following logistic model:where pi is the probability that individual i is affected , ci is an integer that indicates the number of rare CNVs that an individual i has , si is the average size of those events , gi is the count of gene within a pre-specified gene set affected by a cnv , and e is an error term . The terms θ , γ , β0 , and β1 are logistic regression parameters that are optimally determined to maximize the likelihood of the data . The θ term ( the intercept ) represents the background log likelihood for each individual , γ is the increase in log-likelihood per affected gene within the gene set , β0 is the increase in log-likelihood per rare CNV , and β1 is the increase in log likelihood per kilobasepair of average rare CNV size . The cnv-enrichment-test simply tests if γ is significantly different from 0 . In principle , previous studies in schizophrenia that have shown excess CNVs in affected individuals corresponding to a positive β0 . It has also been demonstrated that individuals with neuropsychiatric disease often have larger events , consistent with a positive β1 term . On the other hand , if there is a “causal” gene set g , then adding it to the model should attenuate the magnitude of both β0 and β1 and result in a convincingly positive γ . An independent odds ratio estimate , eγ , can be calculated for the additional increased risk of disease if an event affects a gene in set g . This approach is not confounded by functionally related genes that cluster on the genome . Since risk is estimated on a per individual basis , a single spurious observation will not dramatically impact the statistical significance of any of the parameter estimates . So , a rare single event , which happens to overlap multiple related genes within the gene set that is being tested , will not contribute substantially to the significance of γ - even though potentially many genes from that pathway are implicated . Of course , if many such events are observed , with a proclivity towards either cases or controls , then estimates for γ might appropriately be more significant . The approach can be extended to do a meta-analysis if patient data is aggregated , and indicator variables are included to denote the dataset that the patient sample was derived from . Indicator variables would potentially account for specific differences across data sets , such as the proportion of individuals that are cases and also underlying biases in case severity . This approach can be facilely applied to gene-sets ranging widely in size . It can equally be applied to a single gene , for example to identify whether a gene such as NRXN1 has more case-events than control-events after controlling for genome-wide differences in CNV size and rate . It can also be easily applied to the set of all genes in the human genome to test if genes in general are more often affected in cases than controls . We caution that in data sets with too few individuals , association to smaller gene sets might be difficult to detect given power limitations; furthermore the asymptotic p-value might be inaccurate . In cases where too few events have been genotyped the asymptotic p-value can be replaced by a p-value based on robust permutation testing instead . We have implemented this test in the publicly available genetic data analysis software , PLINK [23] . To demonstrate that the cnv-enrichment-test does not detect spurious associations due to gene features that predispose key gene sets towards CNVs , we carefully considered gene size . We created an extreme hypothetical scenario ( S0 , see Table S2 ) . Here , every fifth gene was designated as a hypothetical “brain gene”; brain genes were set to be considerably larger than other genes ( 50 kb versus 10 kb ) . For a single hypothetical chromosome , 250 Mb in length , we placed 2000 evenly spaced , non-overlapping genes . In all scenarios we simulated CNV data for 2000 cases and 2000 controls , specifying the mean CNV size at 100 kb ( range 10 kb to 150 kb , standard deviation 30 kb ) and the CNV rate per individual at 0 . 25 . Reassuringly , in this simulation cnv-enrichment-test for “brain-genes” demonstrated p<0 . 05 association in 4 . 1% of 10 , 000 simulated datasets , suggesting that it estimates the type I error rate accurately ( see Figure 1 ) . However , in practice , differences between the size and rate of CNVs might be present due to true genetic differences between cases and controls , as demonstrated in neuropsychiatric disease , or technical differences in array intensity or genotyping platform . Our method must be robust to these differences and must not spuriously identify pathways with large genes as a consequence of these differences . To test for this we created four extreme scenarios ( S1–S4 , see Table S2 ) . Under S1 , we dramatically reduced the control rate of CNVs to 0 . 05/individual while retaining the same rate in cases ( 0 . 25/individual ) . Under S2 , we fixed the rate at 0 . 25/individual in both cases and controls , but reduced the mean CNV size in cases ( 60 kb ) compared to controls ( 100 kb ) . Under S3 , we assigned cases the greater rate and mean size ( 0 . 25/individual and 100 kb ) compared to controls ( 0 . 05/individual and 60 kb ) ; this scenario is analogous to schizophrenia where events are larger and more frequent in cases . Under S4 , we assigned cases had a greater rate , but smaller mean size ( 0 . 25/individual and 60 kb ) compared to controls ( 0 . 05/individual and 100 kb ) ; this scenario might occur if higher quality genotyping is applied in cases only resulting in better ability to detect smaller CNVs than in controls . We found that the proposed method that controlled for both CNV rate and average CNV size was robust under each of these extreme scenarios and for 10 , 000 simulated datasets demonstrated appropriate type I error rate at p<0 . 05 under all scenarios ( see Figure 1 ) . To illustrate the importance of controlling for CNV rate and size in this setting where a pathway consists of systematically larger genes , we examined more limited models that do not control for either or both the CNV rate and size . All of these models caused inappropriately high type I error rates under at least one of the above scenarios ( see Figure 1 ) and would demonstrate spurious association to “brain genes” . A simple association test ( M0 ) that does not account for either for CNV rate or size at all demonstrates higher rates of false associations under all simulated scenarios where there are case-control differences in size and rate of CNVs ( S1–S4 ) . Similarly , controlling for differences in rate only ( M1 ) demonstrates higher rates of false associations under almost all simulated scenarios , except for S0 and S1 . Controlling for differences in size only ( M2 ) demonstrates higher rates of false associations under almost all simulated scenarios , except for S0 and S3 . Finally , controlling for differences in total CNV burden ( M3 ) demonstrates higher rates of false associations under S3 and S4 all simulated scenarios . To broadly define genes that control brain function , we used a gene expression tissue atlas to define a broad set of 2 , 531 preferentially brain-expressed genes ( see Materials and Methods ) . For secondary analyses , we compiled three more sets of general interest to neuropsychiatric disease: ( 1 ) 455 neuronal-activity genes defined by Panther and highlighted previously in schizophrenia by Walsh et al [1] , ( 2 ) 126 learning genes defined by Ingenutiy and highlighted by Zhang et al in bipolar disease [8] , and ( 3 ) 209 synapse genes defined by Gene Ontology . The gene sets overlap; 12 genes are in all four sets . To demonstrate some of the limitations associated with standard set enrichment tests to assess critical gene functions examined the aforementioned gene sets in rare CNVs from controls recruited from the general populations . We used Affymetrix 6 . 0 chips in conjunction with stringent and uniform quality control to genotype 2 , 415 unaffected individuals ( see Table S3 and Table S4 ) from four separate studies [8] , [24]–[26]; hereafter referred to as ‘meta-controls’ . We identified 1 , 054 single event deletions ranging from 20 kb to 1 . 9 Mb in size . To obtain the most confident calls possible , we focused only on deletions ( see Materials and Methods ) – though including duplications does not substantially impact our results . Strikingly , many of the genes that are disrupted ( and therefore also overlapped ) by rare deletions within the meta-controls have been proposed as candidate genes for neuro-developmental diseases including: GRM5 , GRM8 , FHIT , OPCML , PTPRD , NRXN3 , NRG3 , CNTNAP2 , AUTS2 , CTNNA3 , DLG2 , ERBB4 , PTPRM , and NRXN1 . All of these genes are among the largest in the human genome , with transcripts extending from 550 kb to 2 . 2 Mb of genome . Except for GRM5 and PTPRM , they are all greater than 1 Mb in length . In particular DLG2 , ERBB4 , PTPRM , and NRXN1 were disrupted by 12 individual events in our study; Walsh et al . highlighted these four genes as potentially pathogenic based on pathway analysis [1] . As previously observed by Redon et al [14] and Yim et al [15] , genes affected by rare CNVs are involved disproportionately in brain function in this control population . The set of genes disrupted by deletions within the meta-controls are enriched for brain-expressed genes ( OR = 2 . 0 , p = 2×10−8 ) and other brain function gene sets as well ( see Figure 2 ) . The enrichment is present , though somewhat less pronounced , if all genes overlapping deletions are included ( OR = 1 . 63 , p = 4×10−6 , see Figure 2 ) . To explain this enrichment of rare CNVs affecting brain-function genes in controls , we conjectured that the gene set enrichment approach is confounded by gene size . Three observations support this possibility . First , the transcripts of brain-expressed genes are significantly larger than of other human genes ( p = 9×10−82 by non-parametric rank-sum test , see Figure 3A ) . The median length of all human gene transcripts is 28 . 2 kb; in contrast the median length of brain expressed gene transcripts is 47 . 2 kb ( 1 . 7 fold longer ) . In fact of the genes longer than 1 Mb , 32 out of 48 ( 67% ) are brain-expressed . Genes in the three other gene sets are also significantly longer ( 1 . 2–3 . 1 fold ) . Second , we note that the genes affected by CNVs are also large . Genes disrupted by events in these meta-controls , as well as previously published data sets by Zhang et al , Walsh et al , and the ISC were large ( p<2×10−10 , see Figure 3B ) . The bias towards large genes is still present , though mitigated , if the analysis is expanded to include all overlapping genes ( p<0 . 01 , see Figure 3B ) . Smaller genes overlapping a CNV are much more likely to be fully contained by that CNV while larger genes are more likely to extend beyond the boundaries of the CNV and hence be disrupted by that CNV . Third , almost all gene ontology [13] ( GO ) categories consisting of genes with an average size >200 kb are preferentially affected by rare deletions within the meta-controls ( see Table S5 ) . These codes implicate functions such as cell adhesion and recognition , neuron recognition , and synaptic pathways . To quantify the extent to which observed enrichment for these gene sets was simply a consequence of their large size , we tested whether randomly placed genomic segments affect genes with brain function genes preferentially also ( see Table 1 ) . We created 1 , 000 sets of 1 , 054 randomly positioned non-overlapping segments of equal size and probe density as those rare deletions observed in the meta-controls ( see Materials and Methods ) . Brain-expressed genes were enriched among overlapping genes [OR = 1 . 67 ( 1 . 17–2 . 26 ) ] and disrupted genes [OR = 2 . 08 ( 1 . 71–2 . 49 ) ]; the enrichment for brain expressed and other brain genes sets was comparable to the enrichment in observed data . However , there are two key differences in the results of real rare CNVs and simulated CNVs . Observed rare deletions overlap 35% fewer genes than random segments – suggesting unsurprisingly that deletions overlapping genes are selected against . Possibly , events affecting potentially critical genes that , if affected , disrupt normal human development are selected against . But , on the other hand , the pattern for the largest genes is strikingly different – the observed rare deletions actually overlap 26% more of those genes >1 Mb in length than random segments . This suggests a predilection for large genes that cannot be accounted for simply by their larger genomic footprint . To explain the discrepancy between the size and number of genes affected in real CNVs and simulated segments , we speculated that while rare events affecting genes are negatively selected against , those that affect large genes might be less strongly selected against . Possibly , large genes have certain structural features that tend to make them relatively preferred targets of rare CNVs above and beyond their simple large size . For example , a CNV within a long gene might be more likely to fall within a large intron and not disrupt the coding sequence , and therefore have less-clear relevance to gene function . Furthermore , since genes tend not to overlap , a CNV of a particular size that overlaps larger genes may affect fewer genes than one that overlaps many smaller genes , and may therefore be less likely to impact some nearby essential gene . To test whether these factors might play a role we tabulated three relevant structural features for each gene ( see Materials and Methods ) : ( 1 ) transcript length , ( 2 ) a gene neighborhood density score , representing the expected number of additional nearby genes that a randomly placed CNV affects , and ( 3 ) a gene structure score , that represents the expectation that a randomly placed overlapping deletion is fully intronic . The first parameter simply accounts for the size of the ‘target’ . The other two parameters account for the possibility that CNVs overlapping certain genes might be more likely to be functionally consequential . We found that all of these variables individually correlated with the likelihood that a gene is overlapped by deletions in meta-controls ( see Figure S1 ) . We then conducted a conditional analysis and found that even though they are inter-correlated , they each independently predict the probability that a gene is deleted in the meta-controls – removal of any single parameter significantly affects a logistic regression model's predictive ability ( see Table 2 ) . The additional factors of gene density and structure could account for the reduced number of affected genes overall and the increased proportion of larger genes compared to random segments in the genome . One possible strategy to correct gene set based analyses is to devise a score that encapsulates the structural features of genes , and their predicted propensity to be affected by a CNV . This provides a robust approach to assess pathway enrichment in he suboptimal situation when controls are not available ( e . g . when evaluating a collection of de novo case-only Autism deletions ) . We present such a CNV-propensity gene score ( CNVprop ) that represents an empirical estimate of the log-likelihood that a gene is overlapped by a CNV based on gene structural features based on the parameters from Table 2 . CNVprop can be used as a covariate within a logistic regression framework in assessing enrichment of a gene set . We provide the CNVprop scores of genes in Table S6 . While other methods to correct for gene size have been proposed in the literature , they do not specifically account for additional effects from gene density and intron structure , which are likely specific to CNV events . This approach , however , is still not ideal since it fails to account for multiple genes contributed by a single event , or genes being affected multiple times by an individual CNV event . To further demonstrate application of gene set analysis and its potential pitfalls , we used a large data set published by the ISC with many rare ( <1% frequency ) deletions and duplications identified from 3 , 391 affected by schizophrenia and 3 , 181 unaffected individuals ( see Table S3 ) . In order to replicate the analysis published by Walsh et al [1] , we conducted set-based analyses of genes disrupted by CNV events within the ISC cases . We observed enrichment of brain-expressed genes ( p = 3×10−11 , two-tailed Fisher's exact , Figure 2 ) . However , when we examined genes disrupted within controls in the ISC , we observed similar evidence for brain-expressed genes ( p = 3×10−10 , see Figure 2 ) . Critically , the odds ratios ( ORs ) for enrichment of brain-expressed genes among genes disrupted in affected individuals and unaffected individuals were difficult to distinguish in this analysis . We observed similar trends towards enrichment for brain-expressed genes overlapped by CNVs ( OR = 1 . 1 , p = 0 . 06 for cases , OR = 1 . 08 , p = 0 . 20 for controls , data not shown ) and overlapped by very rare single event CNVs ( OR = 1 . 5 , p = 0 . 004 for cases , OR = 1 . 6 , p = 0 . 01 for controls , data not-shown ) . Furthermore , with the exception of the learning genes , all brain function gene sets demonstrated significant enrichment within ISC cases and controls ( see Figure 2 ) . We applied the same analysis to a data set with a small number of CNVs published by Walsh et al and demonstrated similar effects ( see Figure 2 ) . Cases tended to be more statistically significant for all of the gene sets than controls , since they were better powered with more affected genes . However , confidence intervals were wide in this analysis , and it was unclear it there were true case-control differences . In both data sets – while statistically significant enrichment for brain function genes is observed in cases , it is not clear that the effect size is any different than in controls . We applied the case-control cnv-enrichment-test to check CNVs published by the ISC and by Walsh et al to test whether case events were enriched for genes with brain function relative to controls . In the ISC data , we had already reported elsewhere increased genome-wide rates and sizes for case CNVs [2] . Walsh et al had demonstrated genome-wide enrichment separately . We applied the cnv-enrichment-test to the four gene sets ( brain-expressed , neuronal-activity , learning and synapse as described above ) . The results in Table 3 report the empirical 1-tailed p-values for a test of enrichment of the genes in the set relative to the genome-wide baseline rates of all CNVs; for the smaller gene sets standard asymptotic tests yielded unreliable estimates , due to the sparse nature of the data ( for example 7 case events , 0 control events for neuronal genes in Walsh et al ) . In this context , the empirical significance values obtained via permutation will be robust to these sparse cell counts . Of course , for the larger gene sets , and all of the gene sets in the larger ISC data set , analytical p-values corresponded closely to permuted p-values . There was no evidence of enrichment among case-CNVs compared to control CNVs for brain-expressed and synapse genes ( p>0 . 12 , one-tailed analysis , see Table 3 ) . This is in marked contrast to the observed enrichment of these same brain gene sets in the case-only analyses presented in Figure 2 that did not account for gene size . However , the neuronal-function gene set demonstrated evidence of association to schizophrenia cases for both Walsh et al ( p = 0 . 00045 ) and the ISC data ( p = 0 . 04 ) . There was also evidence of association of the learning gene set within the ISC data ( p = 0 . 009 ) but not in the Walsh et al data ( p = 0 . 35 ) . We want to emphasize that these results are not adjusted for multiple hypotheses testing – and the plausible number of independent gene sets . In this study alone we have tested four separate gene sets . Ultimately , convincing associations will require larger data sets . As additional samples are genotyped for CNVs , the relevance of the neuronal-function genes might be more clearly established . Of note , considering only deletions within the ISC data , the effect of neuronal-function gene set enrichment is stronger ( p = 0 . 0067 , with higher rates in cases ) . Similarly , considering only deletions within the ISC data , the effect of the learning gene set enrichment is also stronger ( p = 0 . 002 , with higher rates in cases ) . In both cases neuronal-function and learning gene sets , the effect sizes associated with an event affecting a gene is modest ranging from 1 . 2–1 . 7 . This suggests that even if the set associations are ultimately validated , that rare CNV events affecting genes within these sets certainly do not fully explain the pathogenicity of rare CNVs . The cnv-enrichment-test is an extremely versatile test to identify whether a gene set of interest is associated with case-control status . We have shown that it is robust to confounders , such as case-control differences in CNV rate and gene size , while standard gene set enrichment approaches are not . Since the cnv-enrichment-test can be applied easily to a wide range of gene sets , there may be the temptation to examine data sets by testing a compendium of gene-sets . Generally , we discourage this approach , and urge investigators to look at specific sets of interest . Assessing the significance of association statistics when testing a large compendium of gene-sets is complex since there is a large number of highly overlapping sets; correcting for the large burden of multiple hypotheses testing appropriately can be challenging . However , should one decide to test such a compendium of gene-sets , it is important that investigators permute the case-control status within their own data set , and apply the same battery of tests to make sure that the actual data set is obtaining levels of significance that are beyond that of the permuted data sets . We have also shown that pathway analyses with standard gene set enrichment approaches are confounded by gene size and structure . This issue is of particular importance when considering genes with brain function – since those genes are significantly larger than other human genes . We have demonstrated how a large set of brain-expressed genes seem to be impacted by CNVs in both case and control populations when using gene set enrichment approaches , and how this effect is largely the consequence of the size of these genes . The brain-expressed genes were selected for having significantly greater expression in neuronal tissues as opposed to non-neuronal tissues . Certainly genes with important brain functions that are ubiquitously expressed in all tissues might be missed by such a strategy , as might genes with very low expression levels overall . However , we observed very similar results for three other separately curated sets of genes with brain function; this suggests that gene size and spurious pathway associations may be of particular importance for brain function genes . The approach we describe here can be applied more broadly than within the context of CNVs; the cnv-enrichment-test can be applied to any situation where disease-associated genomic segments are defined . For example , linkage disequilibrium blocks around associated SNPs can be defined as disease-associated genomic-segments . The potential for gene size and structure confounding pathway analyses extends beyond CNV studies , and applies equally to pathway analyses within other types of genetic studies , including SNP association studies , as noted by Wang et al [27] and exon re-sequencing studies . For example , in a study looking at classes of genes that are disproportionately affected by rare exonic mutations , the total length of the coding sequence will be a key confounding variable . Similarly , studies looking at classes of genes that contain a single SNP nominally associated to disease , confounding variables might include the number of independent SNPs examined , the physical size of the gene , and the recombination hotspots across the length of the gene . In any case , careful case-control comparisons are essential to avoid these confounders . Many of the genes involved in brain function are compelling candidate genes for neurological and psychiatric diseases – and indeed they may be the most vulnerable to CNVs . The purpose of this manuscript is not to question the results of the original publications , but to rather set up a rigorous statistical approach that allows investigators to accurately estimate effect sizes of events impacting specific gene sets of interest and also to precisely replicate reported results .
We obtained rare event deletions and duplications from Table 2 in the original publication of the data [1] . Rare ( <1% frequency ) event deletions and duplications were provided directly by request from the International Schizophrenia Consortium . We obtained data from unaffected individuals with informed consent from four Institutional Review Board approved studies: macular degeneration [26] , myocardial infarction [25] , bipolar disease [8] , and multiple sclerosis [24] . We obtained Affymetrix 6 . 0 raw intensity data for all samples and ran the Birdsuite software on each plate individually [31]; CNV calls were based on Birdseye output . We then analyzed healthy unaffected individuals from each of four studies separately . First we filtered individuals on SNP data , removing individuals with >5% missing data . Second , in situations where Birdseye called two nearby segments ( <10 kb ) with identical copy number and there was a low confident segment in between ( LOD<3 ) , we merged those segments . Third , we exclude all CNVs that ( 1 ) overlap CNVs from a map of common variation [32] , or ( 2 ) failed stringent quality control criteria ( <20 kb in length or <10 LOD or <10 probes ) . Fourth we removed those individuals in each study that were outliers in either excessive number of CNVs , or in excessive aggregate length of CNVs – we defined outlier as the median plus the 1 . 5 times the inter-quartile range . We then combined all CNVs into a single data set , and identified single-events ( i . e . non-overlapping ) deletions . We produced 1000 sets of non-overlapping segments throughout the genome . Each set consisted of segments matched for size and probe-denisty ( +/−10% ) to each observed single-event deletions in meta-controls . Since we were simulating rare events , random events were not allowed to overlap regions with known copy number variation [32] or in regions where we observed an overlapping event ( i . e . not a singleton ) in the meta-controls . For each gene we defined three parameters ( 1 ) gene length , ( 2 ) gene neighborhood density score , and ( 3 ) gene structure score . Gene length was simply the length of the gene transcript in mega-basepairs . To calculate a neighborhood density score for a gene , we consider a CNV overlapping a gene . The neighborhood density score is then the expected number of additional nearby genes overlapped by the same CNV . To empirically estimate the distribution of sizes of rare CNVs , we utilized the sizes , s , of observed single event deletions in the meta-controls . Then to calculate the gene neighborhood density score , gdi , for gene i , we used the following formula:where p is a genomic position , overlapi ( p , s ) is an indicator function that is 1 if a segment of length s starting at position p overlaps gene i , or is otherwise 0 . Similarly overlapall ( p , s ) is the number of genes that a segment of length s starting at position p overlaps . In the numerator we subtract one off , since we want to exclude gene i itself To calculate a gene structure score , we calculated the expected proportion of overlapping CNVs that would not affect the coding sequence of the gene ( i . e . be fully intronic ) . To empirically estimate the distribution of sizes of rare CNVs , we again used the sizes , s , of observed single event deletions in the meta-controls . Then to calculate the gene structure score , gsi , for gene i , we used the following formula:where p is a genomic position , overlapi ( p , s ) is an indicator function that is 1 if a segment of length s starting at position p overlaps gene i , or is otherwise 0 . Similarly intronici ( p , s ) is an indicator variable that is 1 if a segment of length s starting at position p does not overlap a coding sequence , or otherwise is 0 . In order to produce a framework to test gene-sets and their association to disease , we used a linear/logistic regression framework in which phenotype is regressed on the number of genes intersected ( or disrupted ) by one or more CNVs and covariates . We considered five different models to test for enrichment of CNVs in a pathway of interest , and tested them with simulated datasets . For a disease outcome , a standard model , M0 , is as follows:where for individual i , gi is the number of genes in a pathway of interest that intersected/disrupted by a CNV . The θ term is the logistic regression intercept and represents the background log likelihood for each individual , while γ is the logistic regression parameter for gi . Model M1 controls for potential genome-wide differences in CNV burden between cases and controls:where ci is the total number of CNVs in a given individual i . The β0 term is the logistic regression parameter for ci . Model M2 alternatively controls for CNV size:where si is the individual's mean CNV size in kb . If for a particular individual ci = 0 ( i . e . they do not have any CNVs ) then si is set to the sample mean of s rather than 0 or missing . ( Otherwise , if many individuals have no CNVs , a strong correlation will be induced between the rate and average size of CNVs . ) The β1 term is the logistic regression parameter for si . Model M3 alternatively controls for an individual's total CNV burden expressed in terms of total kb deleted or duplicated , written here as the product of ci and si:The β2 term is the logistic regression parameter for ( ci·si ) . Finally , the cnv-enrichment-test model controls explicitly for potential case/control differences in both the number and size distributions of CNVs:This is the model introduced in the main text labeled as the “cnv-enrichment-test” . As above , if for a particular individual ci = 0 ( they do not have any CNVs ) then si is set to the mean size of all CNVs in the sample as opposed to zero . Under all circumstances , the null hypothesis for the 2-sided test of enrichment is H0: γ = 0 . We conducted simulations to understand the performance characteristics of these different analytic approaches ( M0–M3 and cnv-enrichment-test ) to test for enrichment of case CNVs in a set of genes . We explicitly adopt extreme conditions in these simulations , to best illustrate the robustness of each approach under the broadest range of conditions . For each individual , we simulated data for a single hypothetical chromosome , 250 Mb in length . We placed 2000 evenly-spaced , non-overlapping genes on the hypothetical chromosome , where every fifth gene was designated as a “brain gene” . We assigned brain genes to be considerably larger than other genes ( 50 kb versus 10 kb ) . In all scenarios we simulated CNV data for 2000 cases and 2000 controls . For cases and controls , the mean CNV size was either 60 kb or 100 kb , as detailed in Table S2 ( range 10 kb to 150 kb , standard deviation 30 kb ) . Under all scenarios , individuals had either 0 or 1 CNV , with rates given in Table S2 . All datasets were simulated under the null hypothesis of no enrichment for brain genes; that is , CNVs were randomly placed on the hypothetical chromosome , similarly for both causal and controls . Under five scenarios , S0 to S4 , we altered the mean CNV rate and CNV size for cases and controls independently , in order to induce enrichment of CNVs in brain genes arising solely as a consequence of CNV rate and size . Under the first scenario , S0 , there were no differences between cases and controls in the rate and size of CNVs: we therefore expected all methods to give appropriate type I error rates here . Under S1 , the rate of CNVs was higher in cases . Under S2 , the average CNV size was smaller in cases . Under S3 , cases had a greater number , and larger , CNVs than controls . Under S4 , cases had a greater number , but smaller , CNVs than controls . For each scenario , we simulated 10 , 000 datasets to calculate the type I error rate for the enrichment test , for a nominal rate of 0 . 05 . This test is implemented in PLINK v1 . 07 ( –cnv-enrichment-test ) . It is appropriate for either continuous or disease traits and allows for the inclusion of multiple other covariates and for empirical significance tests . The following examples illustrate basic usage . If the file genes . dat contains the locations of all genes ( i . e . as available from the resources section of the PLINK website , glist-hg18 ) and the file pathway . txt is a file of gene names forming the pathway to be tested for enrichment and the CNV data are in the files mycnv . cnv , mycnv . cnv . map and mycnv . fam ( see website CNV page for details ) , then one can ask whether a ) genes are enriched for CNVs , b ) a subset of genes are enriched , relative to the whole genome , c ) a subset of genes are enriched , relative to all genes . The latter form of the enrichment test might be desirable , for example , to determine whether any enrichment is general to all genes , or specific to a subset of genes . a ) Enrichment of genic CNVs . /plink ––cfile mycnv ––cnv-count genes . dat ––cnv-enrichment-test b ) Enrichment of pathway genes CNVs , relative to all CNVs . /plink ––cfile mycnv ––cnv-count genes . dat ––cnv-subset pathway . txt ––cnv-enrichment-test c ) Enrichment of pathway genes CNVs , relative to all genic CNVs . /plink ––cfile mycnv ––cnv-intersect genes . dat ––cnv-write my-genic-cnv . /plink ––cfile my-genic-cnv ––cnv-count genes . dat ––cnv-subset pathway . txt ––cnv-enrichment-test The usual modifiers ( to define intersection differently , allow for a certain kb border around each gene , filter on CNV size , type or frequency , etc ) are all available . Under all circumstances , 2-sided asymptotic p-values are returned . Alternatively , permutation testing can be applied and 1-sided empirical p-values are returned ( positive enrichment in cases , based on estimated regression coefficient ) . For additional information consult the PLINK website ( http://pngu . mgh . harvard . edu/purcell/plink/ ) , the resources subsection ( gene list ) ( http://pngu . mgh . harvard . edu/purcell/plink/res . shtml ) , or the CNV file format subsection ( http://pngu . mgh . harvard . edu/purcell/plink/cnv . shtml ) . | Specific rare deletion and duplication events in the genome have now been shown to be associated with neuropsychiatric diseases such as 16p11 . 2 to autism and 22q11 . 21 to schizophrenia . However , controversy remains as to whether rare events impacting certain pathways as a group increase the risk of disease , and if so , what those pathways are . Other studies have used standard gene-set enrichment approaches to demonstrate that events discovered in cases contain more genes in neuro-developmental pathways than would be expected by chance . However , these analyses do not explicitly compare the relative enrichment in cases to any enrichment that may also be present in controls . Therefore , they can be confounded by the large size of brain genes or by larger size or frequency of CNVs in cases . Here we propose a case-control statistical test to assess whether a key pathway is differentially impacted by CNVs in cases compared to controls . Our approach is robust to skewed gene sizes and case-control differences in CNV rate and size . | [
"Abstract",
"Introduction",
"Results",
"and",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"genetics",
"and",
"genomics/complex",
"traits",
"genetics",
"and",
"genomics/genetics",
"of",
"disease",
"neurological",
"disorders/neurogenetics",
"neurological",
"disorders/neuropsychiatric",
"disorders",
"computational",
"biology/genomics",
"genetics",
"and",
"genomics/bioinformatics",
"genetics",
"and",
"genomics/medical",
"genetics"
] | 2010 | Accurately Assessing the Risk of Schizophrenia Conferred by Rare Copy-Number Variation Affecting Genes with Brain Function |
Sexual contact patterns , both in their temporal and network structure , can influence the spread of sexually transmitted infections ( STI ) . Most previous literature has focused on effects of network topology; few studies have addressed the role of temporal structure . We simulate disease spread using SI and SIR models on an empirical temporal network of sexual contacts in high-end prostitution . We compare these results with several other approaches , including randomization of the data , classic mean-field approaches , and static network simulations . We observe that epidemic dynamics in this contact structure have well-defined , rather high epidemic thresholds . Temporal effects create a broad distribution of outbreak sizes , even if the per-contact transmission probability is taken to its hypothetical maximum of 100% . In general , we conclude that the temporal correlations of our network accelerate outbreaks , especially in the early phase of the epidemics , while the network topology ( apart from the contact-rate distribution ) slows them down . We find that the temporal correlations of sexual contacts can significantly change simulated outbreaks in a large empirical sexual network . Thus , temporal structures are needed alongside network topology to fully understand the spread of STIs . On a side note , our simulations further suggest that the specific type of commercial sex we investigate is not a reservoir of major importance for HIV .
Spatiotemporal heterogeneities in sexual contact patterns are thought to influence the spread of sexually transmitted infections ( STIs ) . Since epidemics can be a society-wide phenomenon , and sexual contact patterns can have structure at all scales , we need population-level sexual network data to understand STI epidemics . Unfortunately , it is hard to collect sexual contact data on that large a scale . Instead , people have focused on small-scale studies using interviews [1]–[3] or contact tracing [4]–[8] , or they have studied larger sample sets using random sampling surveys [9]–[13] . Small surveys and contact tracing risk missing large-scale structures [13] and emergent phenomena . Large-scale surveys , on the other hand , have mainly collected the number of partners , but not the connections between them . An alternative way of gather information about sexual contact patterns , which covers a large number of people and explicitly maps their connections , is to use Internet data . In our study , we used a dataset of claimed sexual contacts between Brazilian escorts ( high-end prostitutes ) and sex buyers [14] . Contact patterns of commercial sex cannot be generalized to a whole population , but they do contain relevant information that can be used to study possible transmission pathways within a social group . Our dataset has information about the time and location of sexual contacts covering six years and 16 , 748 individuals . Sexual contact patterns have temporal correlations both at an individual and at a population level [14] . Much like the network structure , temporal structures may influence epidemics in several ways . For example , consider three individuals , A , B , and C , where B and C are in contact first , and later A and B . Considering the temporal order of the contacts , disease cannot spread from A to C via B , but in a standard static network representation this sequence of events is lost , so C appears reachable from A via B [15]–[19] . A conspicuous temporal structure in human behavior that we also observe in our data is bursts of activity during which people are very active for a limited period at a time [20] . Another example of a temporal structure is the long-term behavioral change in which new individuals enter the system and others leave . These temporal effects result in a heterogeneous distribution of inter-event times [14] . To investigate such temporal effects empirically requires time stamps on the contacts . Internet data sets like ours , as opposed to most above-mentioned data , contain just such information . Extending epidemiological models to include space is a common step towards inclusion of structure beyond the well-mixed assumption [21] , [22] . Geography leaves several imprints on the contact structure and thus on disease spread , making the contact network larger than a random graph in terms of graph distances; it also creates the network clusters corresponding to densely populated areas [14] . These effects stress the importance of network-data sets covering a wide geographic area . Our data set , although just a small fraction of the global sexual networks , probably represents a substantial fraction of the Internet-mediated escort business of Brazil [14] . In this paper , we address the question of how the dynamic contact structure in the contact data of Rocha et al . [14] affects epidemic spread in general . For most of our study , we look at spread processes confined to our contact data . Because of the lack of similar data on other types of sexual interaction , it is hard to draw conclusions about the role of the escort business on the spread of STIs in society as a whole . Rather , we investigate the contribution of topological , temporal , and geographic structure on transmission pathways within this specific type of commercial sex . We do , as an example of how our study can be applied to more specific cases , make a crude estimate of the role of our commercial-sex network in the spread of HIV in a population-wide context .
The web community from which our dataset is obtained is a public online forum openly visible online . The full dataset is available as support information ( Dataset S1 ) . It is oriented to heterosexual males ( sex buyers ) , who evaluate and comment on their sexual encounters with female prostitutes ( sex sellers ) , both using anonymous aliases . The posts on the forum are organized by the city location of the encounter and by type of prostitution as defined by price level and mode of acquiring customers ( for example , escorts , street sex-workers , brothels ) . We focus on the escorts section , the most expensive form of prostitution [23] of the forum , mostly because it is better organized than the other sections—each escort is discussed in a unique thread . This forum can straightforwardly be represented as a bipartite network—we connect a sex buyer ( one type of node ) posting in a forum thread to the escort ( another type of node ) discussed in the thread . An edge in this network represents one sexual encounter between two individuals . The edges are tagged with the dates of the posts , which we take as an estimate of the time of the sexual encounters , even though the sex buyers often post about several encounters at the same session . Consequently , the order of the posts does not have to be exactly the same as the order of the actual encounters . The dataset covers the beginning of the community , spanning the period September 2002 through October 2008 . All in all , 50 , 185 contacts are recorded between 6 , 642 escorts and 10 , 106 sex buyers . Even though the network is spread out over twelve Brazilian cities , these contacts make up a network with a largest connected cluster covering more than 97% of the individuals ( see Rocha et al . [14] for a thorough analysis of this sexual network ) . To minimize finite-size effects , we discard the initial 1000 days available in the original data set that correspond to a transient period with fewer users and sparse encounters . One thousand days is an adequate choice , since after that period , the average temporal profile is approximately stationary ( see Ref . [14] for details ) . For better statistical significance , we sample several windows of 800 days . For example , one network sample ( of the original network ) is obtained by taking all nodes and links that occurred in the period between 1000 and 1800 days; another sample is from the period between 1001 and 1801 days , and so on up to the interval 1200 and 2000 days . The average number of vertices of all windows is N = 10 , 526±145 , and the number of contacts ( links ) is C = 27 , 973±3 , 612 , where ± corresponds to the sample standard deviation . Apart from the anonymous aliases of sellers and buyers and time stamps , posts also include the buyers' grades of the escorts' performance and information about the types of sexual activity performed during an encounter , divided into three categories: oral sex ( with or without condom ) , mouth kissing , and anal sex . All posts , however , are assumed to report vaginal intercourse ( random inspection supports this assumption ) . In our simulations , for the sake of simplicity , unless otherwise stated , we use all available links and disregard the fact that they possess different levels of risk . Most contacts between a seller and buyer happen only once . By inspection , several users report that next time they buy sex , they prefer a different escort , even if the encounter was graded good . We can expect that not all Brazilian escorts and customers of such are present in the data . Furthermore , posting about an encounter is a low-cost action by the sex-buyer that gives him status in the community , which makes it likely that the reports from most users are quite complete . For most of this paper , we ignore this and study disease spread on a network defined by our data set as it is , which limits our conclusions to effects of temporal structure relative to various other scenarios . A common method of studying correlations in empirical contact data is to compare a network with ensembles , where some properties ( like the number of nodes and their degrees ) are kept constant and the rest is randomized . In the randomized network versions used in this paper , we conserve the bi-partite structure of the heterosexual network and the number of contacts of each individual . Diverse network structures can affect disease spread [24]–[26]—one example being clustering ( a high density of triangles ) . Our network has a large number of 4-cycles ( the shortest cycle in a bipartite graph ) , and a pronounced community structure , probably a result of the system being geographically embedded [13] , This effect can be studied by randomizing the contact pairs in such a way that we choose two links randomly and swap the respective sex-buyers ( we call this new null model random topological , RT ) . We do not alter the time stamps of the links; hence , the time order of the escorts' contacts is preserved . To remove temporal correlations , we choose two links randomly and only swap their time stamps such that the new encounter time is unrelated to the original , but the network structure is conserved ( this model is named random dynamic , RD ) . Finally , we make a third randomization , where both the temporal and network structures are removed by swapping the time stamps and contact pairs simultaneously ( we call this model random dynamic topological , RDT ) . To put our results in the context of other levels of epidemiological modeling , we also consider two other contact models—a static network approach and the dynamic network model by Volz and Meyers [18] . The static network ( SN ) approach considers the network of pairs , with at least one contact over a time interval of 800 days , and assumes that contacts can happen with equal probability over all these links . This approach is common to most network epidemiological studies ( e . g . , refs . [2] , [3] , [10] ) . To compensate for the removal of the time stamps , we assume that each link has a certain probability of being active . This probability is derived from our original network and depends on the number of contacts C = 27 , 973 and number of different partners K = 21 , 813 in the window of T = 800 days . Thus , the chance of having a contact active is , for our data , pactive = C/KT = 1 . 603×10–3 . The idea of Volz and Meyers's model is that vertices change partners with a probability pchange while keeping the number of partners fixed over time . This model assumes that a vertex is always connected to someone else; however , in our network , in the interval of 800 days , several vertices have only one or few days during which a connection is active . This means that most of the time they are not in a position to catch a disease . To compensate for this effect , and to allow direct comparison to the simulations on the empirical network , we modify the Volz–Meyers ( VM ) model to capture the brevity of partnerships in the data . In our formulation of the VM network , each vertex has a chance pk ( proportional to the original number of contacts of the vertex ) of being active per day . This assures that over the course of 800 days , each vertex has the same number of contacts as in the original empirical network . For each day , we connect pairs of active vertices randomly ( if the number is odd , the remaining active vertex is connected the next day that another active vertex is available ) . Thus , this network has no temporal correlations . We generate VM graphs with 10 , 526 vertices—the same as the average number of vertices in the sampled windows discussed above . We obtain the degree distribution from these sampling windows as well and use it to calculate pk . One can model the spread of sexual infection in various ways to capture the various characteristics of pathogens and contact patterns , and also to serve different aims of explanation and prediction . We explore the effects of temporal correlations on different levels of epidemic modeling . The first disease-transmission model we consider is the Susceptible–Infected–Removed ( SIR ) model . Where all individuals are initially susceptible; upon contact with an infective , a susceptible becomes infective with probability ρ ( probabilities are , unless otherwise stated , per-contact probabilities ) , and after a fixed time δ , a susceptible changes to the removed state . If δ is larger than the vertex lifetime in the network , we get the limit case known as the Susceptible–Infected ( SI ) model . In a static network of finite size and non-zero transmission rate , all vertices will eventually become infected in the SI model . This is not necessarily the case in a temporal network , which makes the SI model more realistic in temporal , compared to static , contact networks . To simulate these models in our empirical network , we first map the sampled network onto a time-ordered list . Each entry in the list is one pair of vertices and the time of the contact . Different contacts between the same pair appear as different entries in the list . Then we divide the list into intervals of 800 days each , as mentioned above . The pairs are ordered according to their times of contact . We select the sex-seller of the first contact of an interval as a source of infection and go through the ordered list infecting a susceptible vertex in contact with an infective vertex with probability ρ . The state of the vertex is updated at each new contact . A way of modeling the fact that the network is connected to a background of sexual contacts would be to include multiple sources of infection . To keep the simulations simple , however , we leave this for future studies . Since our temporal information has a resolution of one day , we do not know the order of contacts within a day . To remove this potential bias , we randomize the order of contacts within a single day 100 times . In line with other studies , and to simplify the model , we assume that both infection and removal ( after time δ ) are immediate , and the transmission probability is constant . The SI model is adequate for modeling the early phase of an outbreak over shorter time scales than the duration of the disease . SIR , on the other hand , is appropriate for simulating diseases having a well-defined infectious stage followed by immunity . As an example , we will investigate HIV at a more detailed level than simply SI or SIR . Hollingsworth , Anderson , and Fraser [27] devised a model for HIV-1 infection with a susceptible stage followed by four distinct infective stages of different infectivity—one acute infection of high infectivity ( over a time-scale of months ) followed by a chronic stage ( lasting for years ) , and another high infectivity stage ( some weeks ) followed by zero infectivity before death . Since our dataset covers only 1000 days , we can omit the last two stages and arrive at a model characterized by an acute stage of transmission probability ρ1 lasting for a time T1 , and a chronic stage of transmission probability ρ2 . We refer to this as SI1I2 model . Strictly speaking , the transmissibility of HIV-1 also depends on gender and other factors such as type of sex and the fact that the viral load transmitted per-contact can spike during the chronic phase because of comorbidities , among other things . A yet more detailed model could also include an age-stratified population , as young infectives tend to influence an outbreak more . Because they are in the network for longer times , they have higher chance to establish more contacts and contribute to transmit the infection [28] . We follow a similar procedure as above to simulate disease spread in the SN and VM networks . For the initial conditions , however , since the probability of being infected should increase with contact rate in case of the empirical networks , we now select the source of infection randomly ( for each realization ) and proportionally to the number of contacts of the vertex . This procedure compensates for the fact that in the empirical network , high degree nodes are necessarily selected more than once as a source of infection . This is because , on average , the chance of an individual's being active at a certain moment is proportional to that individual's number of contacts . The state of the vertex is updated after all vertices have been considered . We run the algorithm 30 , 000 times to obtain averages for these models . A key quantity is the fraction of infected vertices Ω ( the outbreak size ) . If the time evolution is not explicitly stated we refer to Ω at the end of the sampling time window ( 800 days ) . We also run simulations 50 times over different initial conditions to calculate the average values .
A straightforward way of investigating the effects of the temporal and topological structure of contact patterns is to remove different types of correlations by randomization ( see Section The network models ) . In Figure 1 , we investigate effects of the time ordering of contacts by using the SI model with ρ = 1 and compare the simulated epidemics in the original network with the epidemics in the three different randomized versions of it . In Figure 1A , we see that an infection spreads much more slowly in the RD network model , reaching fewer than 50% of the individuals compared to more than 60% in the original network . Thus , correlations in the order in which the contacts occur speed up disease spread . More concretely , one such tendency is that individuals tend to be intensely active over a period of time followed by idle periods . When the time stamps are randomized ( RD model ) , this tendency disappears such that the presence of individuals in the system is now , on average , longer and the contacts less frequent . The average time , between an individual's first and last active period of , increases from 170 . 9±0 . 1 days in the original network to 337 . 5±0 . 1 days after randomization . In addition to correlations in the temporal order of contacts , the topology of the sexual network can also influence epidemics [3] , [6]–[9] , [18] . In Figure 1B , we compare the evolution of epidemics in the empirical network with the RT network model . The evolution of the fraction of infected individuals 〈Ω ( t ) 〉 seems to grow slowly , at least during the initial 200 days; afterwards , the topologically randomized network yields more rapid and pervasive outbreaks ( Figure 1B ) . The more rapid initial epidemic spread in the original network results from the high clustering of contacts within cities . Finally , considering both the temporal and topological information randomized ( RDT model ) , the curve ( evolution of the epidemics , Figure 1C ) is in between those of Figure 1A and Figure 1B . The fraction of infected vertices increases slowly during the initial 300 days , but not more slowly than in the RD scenario in Figure 1A . Later it increases more rapidly and by the end of the sampling period reaches about 70% of the individuals ( a little less than in the RT scenario in Figure 1B , but still , larger than in the original network ) . The limit of high transmission probability ρ = 1 does not reflect actual STI contagion; more realistic values lie in the range 0 . 001≤ρ≤0 . 3 [27]–[28] . In Figure 2 , we present 〈Ω〉rel = 〈Ωρ〉/〈Ωρ = 1〉 , the average number of infected vertices ( for probabilities ρ ) relative to the number of infected vertices when the maximum transmission probability is used ( ρ = 1 ) . The relative number of infected vertices decreases within the initial 100 days and afterwards reaches a minimum for higher transmission probabilities while continuing to decrease slowly for lower rates . The minimum , which corresponds to the time lag of secondary infections , is more pronounced for lower ρ-values . The fact that the curves are fairly constant for times longer than 200 days , that is , that they converge to limiting values , is an indication that our results for the ρ = 1 case hold for other transmission probabilities as well , that is , the time-ordering effects are stronger than the fluctuations from the stochasticity of the contagion process . For lower ρ-values , the curves decrease monotonically , which indicates the existence of an epidemic threshold somewhere between , ρ = 0 . 01 and ρ = 0 . 001 , which we investigate more cautiously below . We note that there is a large diversity of outbreaks even for ρ = 1 . In Figure 3 , we measure the probability distribution P ( Ω ) of outbreak sizes Ω . This , we hypothesize , is a general phenomenon—temporal constraints increase the diversity of outbreaks because they restrict the possible infection paths in the network . There is , however , a local maximum where , for ρ = 1 , a fraction of about 0 . 75 of the vertices gets infected , setting a characteristic outbreak size . This local maximum depends on the transmission probability and decreases for lower values . Another observation is that the outbreak-size distribution becomes less heterogeneous for lower ρ-values . The peak on the very left of the graph indicates that the disease is likely to die out . Note that the network also contains some isolated connected components that , once infected , do not spread the infection to the giant component . To illustrate the effect of different sexual activities , we show the outbreak size distribution for the original network considering only the encounters that involve oral sex without condom , and mouth kissing ( Figure 3B ) . This specific network has roughly the same outbreak-size distribution ( similar shape and scale ) as the original network , despite being about half as dense . Returning to our original network , we investigate the effect of varying ρ , and see that the average outbreak size 〈Ω〉 is an approximately linear function of transmission probability ( see Figure 4A ) . From the figure , it is evident that the epidemic outbreak is practically absent for transmission probabilities lower than about 0 . 19—a de facto threshold effect . Looking in more detail , one can see that this threshold effect is due to the fact that the mean value of large outbreaks vanishes and that the number of small outbreaks increases as ρ→0 ( cf . Figure 3 ) . To investigate whether this threshold value ρ* is an artifact of the finite-size sampling , we use different sampling windows from the complete dataset and see whether the threshold values converge to a common value for different starting points in the dataset . The threshold values should converge to a limit value as the network structure trends toward a steady state . Each window represents the same duration of time ( 800 days ) , but simply starts at a different time ( T0 ) in the original data set . We take the crossing point between the fit of the fraction of infected vertices as a function of ρ to a line and the line of zero secondary infections as an estimate of the threshold value . We see in Figure 4B the apparent convergence of the threshold estimates to values at about ρ* = 0 . 19±0 . 01 for increasing T0 , which is our estimated threshold value for this contact pattern . This threshold seems slightly smaller for the RDT , but significantly smaller for the SN and VM models ( Figure 5A–C ) . We plot the average outbreak size 〈Ω〉 as a function of the duration of the infective stage δ in Figure 6A . Here , we assume the maximum transmission probability ρ = 1 . We proceed to identify estimated threshold values by performing fits of second-order polynomials to the fraction of infected individuals and identify the crossing point with the zero secondary infection line . Performing a similar analysis as for the SI model's transmission probability threshold , but now for the duration of the infective state , we find that the δ-threshold converges to δ* = 31±1 days . Now , fixing the infective stage to δ = 91 days , which is roughly 3 months and well above our estimated threshold of δ* , we perform SIR simulations for different transmission probabilities and compare the outbreak sizes by using the original network , the randomized version ( RDT ) , a static ( SN ) , and a dynamic network ( VM ) ( Figure 5D–F ) . For all cases , the thresholds are above ρ* = 0 . 2 , and the final outbreak size is always larger for the empirical network , suggesting that the temporal correlations , the essential difference between the raw empirical contact patterns , and the models accelerate transmission . Now we turn to the results of the SI1I2 simulation of HIV spread . We fix the acute infective period at T1 = 91 days and study some different combinations of estimated transmission probabilities available in the literature for different societies by using lower ( ρ1 = 0 . 005 and ρ2 = 0 . 0005 ) and higher ( ρ1 = 0 . 01 and ρ2 = 0 . 001 ) bounds [27] , [29] . In Figure 4A , we see that the threshold transmission probability of the SI model is higher than all these values , so we already know that the SI1I2 model on the actual data is below the epidemic threshold . In Figure 7 , we plot the average time evolution of the outbreak size for both our empirical temporal network ( Figure 7A ) and the RDT contact model ( Figure 7B ) . The average outbreak sizes are , as expected , very low ( a fraction of about 10–5 of the population ) for both these contact patterns . For the RDT model and ρ1 = ρ2 = 0 . 01 , the system is just above the epidemic threshold , as can be seen by its convex curve in Figure 7B . A conspicuous temporal feature is that , for the empirical network , the effect of a larger transmission probability of the chronic infection is very small—the ρ1 = ρ2 = 0 . 01 and ρ1 = 0 . 01 , ρ2 = 0 . 001 curves are almost congruent . For the RDT contact structure , the more homogeneous temporal pattern allows the chronic infection to play a greater role , so these two curves diverge after about 200 days , which is about the average interval between two consecutive contacts .
We simulate the spread of infection in what is probably the largest network of self-reported sexual contacts yet recorded . Our data come from a web community of sex buyers who discuss their encounters with escorts . Although the network is spread out over twelve cities , it is to a large extent connected so that a disease could spread from most parts of the system to most other parts . As with any result based on a subset of a network , we should be cautious about extrapolating our results to the entire society , especially since it is hard to compensate for missing links with the information we have . The escorts in our dataset make up about one percent of all Brazilian sex sellers ( of a total of about one million [30] ) . On the other hand , the escorts are a small fraction of all sex-sellers , and we can tell by the way the average degree ( number of partners ) converges that the sampling time is longer than an escort's typical duration in the business [14] . Another complication when it comes to generalizing the results of the paper to a society as a whole is that our sexual network is not an isolated system . It is possible that the infection leaves the community and eventually returns through other individuals . In that case , our model would underestimate the impact of the network in the outbreak . Furthermore , commercial sex is not necessarily driven by the same mechanisms as regular sexual interaction . So , since our data is not comprehensive enough to infer the impact of prostitution on disease spread , we focus on how temporal correlations in the empirical data affect results from random and well-mixed models . From studying the SI model with a 100% transmission probability in our sexual network , we conclude that temporal correlations speed up the epidemics , especially in the early phase of superlinear growth . This effect has important implications both for disease modeling , implying that temporal correlations in contact patterns should not be underestimated , and for intervention methods ( like targeted vaccination ) , where temporal structures could potentially be used to detect important individuals . Furthermore , the temporal effects seem to cause well-defined and relatively high epidemic thresholds , unlike studies of model networks with power-law degree distributions [10] where outbreaks can occur of any non-zero transmission probability . For purposes of comparison , in a finite-sized scale-free network and Susceptible-Infected-Susceptible epidemics with recovery μ = 1 , ρcritical = 〈k〉/〈k2〉 ( where k is the number of different partners of an individual [31] ) , gives ρcritical∼0 . 043 , using our network . The network structure ( apart from the contact-rate distribution ) , on the other hand , slows down outbreaks . Our network has a high density of short cycles and community structure , reflecting the fact that most sex buyers buy sex in one region , presumably their hometown . Both factors , many short clusters and distinct communities are known to slow down diffusion in networks [24]–[26] . Most of our analysis is at a general STI level and indicates that our network is not dense enough to support STI outbreaks for chronic diseases with transmission probabilities lower than ρ = 0 . 19 . The fact that endemic diseases with arguably lower transmission probabilities exist points to the importance of the background sexual contacts . Because of the incompleteness of our data , this does not completely exclude the possibility of escort prostitution as a reservoir of STIs , but it points to a more complex picture . In the support information ( Text S1 ) , in a crude assessment of our dataset contribution to the general STI spread , we suggest that it would only affect the degree-correction of R0 of STIs by a few percent . We also exemplify how temporal structures can affect the spread of a specific pathogen , HIV-1 , by simulations of a refined compartmental model . The simulation results indicate that our empirical network alone cannot sustain an outbreak of HIV-1 . In general agreement with empirical research [32] , [33] , our results suggest that pathways ( like unsafe man-to-man sex , or intravenous drug use ) other than commercial sex are needed to explain the endemic state of HIV epidemics in Brazil [34] . The other studies are , however , from countries other than Brazil; however , they are inconclusive if not controversial [35] , [36] . We believe that the study of temporal aspects of contact patterns is , in general , a promising direction for the future . We intend to investigate how far our conclusions can be generalized to other types of cultures , other forms of commercial sex , and hopefully to non-commercial sexual contact patterns . | Human sexual contacts form a spatiotemporal network—the underlying structure over which sexually transmitted infections ( STI ) spread . By understanding the structure of this system we can better understand the dynamics of STIs . So far , there has been much focus on the static network structure of sexual contacts . In this paper , we extend this approach and also address temporal effects in a special type of sexual network—that of Internet-mediated prostitution . We analyze reported sexual contacts , probably the largest record of such , from a Brazilian Internet community where sex buyers rate their encounters with escorts . First , we thoroughly investigated disease spread in this dynamic sexual network . We found that the temporal correlations in this system would accelerate disease spread , especially at shorter time scales , whereas geographical effects would slow down an outbreak . More specifically , we found that this contact structure could sustain more contagious diseases , like human papillomavirus , but not HIV . These results highlight the importance of prostitution in the global dynamics of STIs . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] | [
"infectious",
"diseases/sexually",
"transmitted",
"diseases",
"public",
"health",
"and",
"epidemiology/infectious",
"diseases",
"physics/interdisciplinary",
"physics",
"public",
"health",
"and",
"epidemiology/epidemiology"
] | 2011 | Simulated Epidemics in an Empirical Spatiotemporal Network of 50,185 Sexual Contacts |
The World Health Organization recommends at least 3 annual antibiotic mass drug administrations ( MDA ) where the prevalence of trachoma is >10% in children ages 1–9 years , with coverage at least at 80% . However , the additional value of higher coverage targeted at children with multiple rounds is unknown . 2×2 factorial community randomized , double blind , trial . 32 communities with prevalence of trachoma ≥20% were randomized to: annual MDA aiming for coverage of children between 80%–90% ( usual target ) versus aiming for coverage>90% ( enhanced target ) ; and to: MDA for three years versus a rule of cessation of MDA early if the estimated prevalence of ocular C . trachomatis infection was less than 5% . The primary outcome was the community prevalence of infection with C . trachomatis at 36 months . Over the trial's course , no community met the MDA cessation rule , so all communities had the full 3 rounds of MDA . At 36 months , there was no significant difference in the prevalence of infection , 4 . 0 versus 5 . 4 ( mean adjusted difference = 1 . 4% , 95% CI = −1 . 0% to 3 . 8% ) , nor in the prevalence of trachoma , 6 . 1 versus 9 . 0 ( mean adjusted difference = 2 . 6% , 95% CI = −0 . 3% to 5 . 3% ) comparing the usual target to the enhanced target group . There was no difference if analyzed using coverage as a continuous variable . In communities that had pre-treatment prevalence of follicular trachoma of 20% or greater , there is no evidence that MDA can be stopped before 3 annual rounds , even with high coverage . Increasing coverage in children above 90% does not appear to confer additional benefit .
Trachoma , caused by ocular Chlamydia trachomatis , is the leading infectious cause of blindness world-wide [1] . In addition to the heavy personal and societal burden inflicted by loss of vision , there are significant economic impacts as well . The 3 . 8 million cases of blindness and 5 . 3 million cases of low vision due to trachoma are estimated to diminish productivity by $2 . 9 billion each year [2] . A multi-faceted strategy to control all phases of trachoma has been endorsed by the World Health Organization ( WHO ) , consisting of Surgery ( to repair lids distorted by trachoma ( trichiasis ) in imminent danger of vision loss ) , Antibiotics ( mass drug ( antibiotic ) treatment ( MDA ) to reduce the community pool of infection with Chlamydia trachomatis ) , Face washing ( to reduce transmission from ocular and nasal secretions ) , and Environmental improvements ( to interrupt transmission and prevent re-emergence ) . Children , particularly pre-school age children , in trachoma-endemic communities are the reservoir of infection and active follicular trachoma [3]–[5] . The WHO has recommended at least three years of annual MDA , with coverage targets of 80% of the population , followed by an impact survey to guide continued MDA . The indication for ongoing MDA is a prevalence of clinical follicular trachoma of ≥10% in children ages 1 to 9 years . However , there are no data on whether increasing coverage in children would result in more rapid declines in infection . Recent data from Tanzania suggest that with coverage less than 80% , more than seven years of annual mass drug administration ( MDA ) might be needed to achieve the target of trachoma less than 5% [6] . On the other hand , one round of MDA with high coverage in low prevalence communities in The Gambia lowered the infection rate to virtually zero although clinical signs of trachoma were still present [7] . Such data suggest that in low prevalence communities , high coverage may obviate the need for three annual rounds of MDA . In addition , relying for cessation of treatment on clinical signs of trachoma may result in unnecessary treatment of communities where infection has been eliminated and only residual disease is present . Yet , in some communities , even low rates of active trachoma are associated with the presence of infection [8] . Research on the optimal use of antibiotic for trachoma control , especially now that MDA distribution may be integrated with other neglected tropical diseases is urgently needed [9] . We hypothesized that increasing the coverage of MDA to greater than 90% as monitored in children would result in more rapid decline in infection and trachoma compared to usual coverage , between 80–90% . Where communities achieved an estimated prevalence of infection less than 5% in children ages ≤ five years ( regardless of status of clinical signs ) , we hypothesized that the community could cease mass treatment without re-emergence . To test these hypotheses , we conducted a community-randomized trial in Kongwa Tanzania .
All study procedure and protocols were approved by the Johns Hopkins University Institutional Review Board and the National Institute for Medical Research in Tanzania . Written informed consent was obtained by parents on behalf of all child participants . ClinicalTrials . gov ID#NCT00792922 . Communities were randomized 1∶1∶1∶1 in a 2×2 factorial design to two different coverage targets: 80%–90% ( usual target ) versus >90% ( enhanced target ) . They were also randomized to two different annual MDA strategies: yearly mass treatment for 3 years ( usual care ) versus yearly mass treatment each year if warranted by the infection prevalence above 5%; otherwise , the MDA would cease for communities in this arm and the community monitored for re-emergent infection ( cessation rule ) [10] . Communities were randomly selected from all communities in Kongwa district based on previous research or assessment suggesting the prevalence of TF was 20% or greater . Within each community , 100 children less than five years old were randomly selected and surveyed for trachoma and infection at baseline and new samples selected every six months for 36 months . Mass drug administration was provided to each community annually , unless the community was randomized to the cessation rule arm and the prevalence of infection or trachoma fell to less than 5% . Our working definition of the cessation rule was if zero children out of 100 in the sentinel sample at 6 months or 18 months had infection , as determined by PCR , then the next round of MDA would not occur . With a sample size of 100 children , finding no infection is associated with an exact upper 95% confidence limit less than 5% . The design of the trial is shown in Figure 1 . By the 18 month survey , no community in the arms randomized to the cessation rule had achieved an estimated prevalence of infection less than 5% , so all communities proceeded to have three rounds of MDA . Thus , for the final analyses , the main effect of the stopping rule was not considered since no communities were stopped . Only the main effect for the coverage arms was analyzed . The eligibility criteria for communities in this trial were as follows: We defined communities as the smallest population level for implementation of services . Kongwa district is part of the National Trachoma Control Program for Tanzania , so all communities outside the trial were offered MDA if the prevalence of TF was >10% in children ages 0–9 years . MDA consists of a single dose of azithromycin , 20 mg/kg up to 1 GM , offered to all residents of the community ages 6 months or older . Infants under age 6 months are offered topical tetracycline . MDA was carried out by a network of paid community treatment assistants ( CTAs ) who were trained and supervised by Kongwa Trachoma Project staff . These were the only staff that had access to the randomization assignment of the community in order to monitor coverage and plan for MDA cessation . The census list of the community was used to monitor coverage , and as each resident presented for treatment , treatment was observed and recorded in the treatment log by the CTA . The numbers of treatment days were scheduled depending on the tally of coverage at the end of the first three days in each community . For subsequent days , the CTAs were notified in advance of persons who missed MDA , and were to go house to house to provide treatment . If , after the first 3 days of MDA , the percentage of children treated was between 80% and 90% for the one coverage arm , or above 90% for the second coverage arm , then treatment for that community stopped . If treatment was below target , then the staff supervisor scheduled subsequent days for the community , with treatment to be held for another day until coverage targets are met or all persons accounted for in the Treatment log . All persons who presented for treatment were treated , even if it meant higher than 90% coverage for communities randomized to 80–90% coverage . Treatment verification was undertaken after each round of MDA by having a sample of five households per CTA re-visited by a supervisor . Each person in the household was queried about receiving treatment . Payment of TSH 1 , 000 per day for the CTA ( $0 . 80 ) was contingent on achieving 80% agreement on coverage of persons in the five households . The primary outcome measures are the prevalence of C . trachomatis infection , and the prevalence of follicular trachoma , at the community level at the 36 months survey ( last follow up December 2011 ) . Using a custom built SAS macro for constrained randomization , the 32 villages were randomly assigned on a 1∶1∶1∶1 to each arm of the trial . This approach reduces the likelihood of a bad randomization outcome by constraining the randomization by baseline trachoma prevalence as a co-variate [11] , thus ensuring balance in each arm . The study statistician had the responsibility for generating the random assignment of communities . After the village leadership had agreed the community would participate , the census and survey completed , the random assignment was provided to the project director in Tanzania . He then informed the MDA implementation team immediately prior to MDA . For each survey , a custom built Access program ( Microsoft Office 2007 ) randomly selected from the most recent census a list of 110 children ages 5 years and under . A random number was assigned to each child and we used the first 110 lowest numbers . Ten children were kept in reserve in case of the unavailability of a child in the first 100 children sampled . Assuming a standard deviation of 0 . 05 within each arm , a correlation of 0 . 5 between baseline and 36 month results , a figure observed from previous studies , and no interaction between factors , we estimate that a total of 32 communities provides greater than 80% power for each main effect . Neither the assumption of plausible variation in a possible interaction effect nor the assumption of a beta distribution rather than a normal distribution altered these estimates to any substantial degree . Although we estimated that between 40% and 80% of communities randomized to the cessation rule arms would cease MDA , in fact zero communities ceased MDA . Thus , for the final analyses , 16 villages were in the 80–90% coverage group and 16 villages were in the >90% coverage group . The survey teams who assessed trachoma were masked to the allocation of the communities in each arm , as they were never shown the allocation and communities were surveyed in no order by treatment allocation . It was theoretically possible that if a community had MDA stopped their allocation to the cessation rule would be unmasked , but none of the communities were stopped . The laboratory at Johns Hopkins University received specimens with labels that could not be linked to persons or study arms by lab personnel , and these were processed for infection masked to intervention . The results were reported only to the statistician and data managers of her team . The community residents who participated in the survey were not told the results of the laboratory findings , since all were eligible to receive azithromycin . Thus , the infection outcomes were double masked .
A comparison of baseline characteristics of each of the 4 groups of communities showed no imbalances in population size , percentage of households with no latrine , percentage more than 30 minutes from water , or average education of head of household . Similar prevalence of follicular trachoma and C . trachomatis infection were observed ( Table 1 ) . For the rest of the analyses , we show only the coverage arms because none of the communities stopped MDA during the study ( Figure 1 ) The treatment coverage at each of the three rounds of mass drug administration was consistently greater in the enhanced target group compared to the usual target group ( Table 2 ) . The average difference between the two groups in terms of coverage was never more than 6% ( third round of mass treatment ) . All of the communities in the usual target arm achieved coverage above 80% in children under age ten years in each of the three rounds . Only at the one year treatment round did one village in the enhanced target group have coverage below 90% . The baseline prevalence of infection with C . trachomatis was not different between the two coverage groups , 20 . 1% and 23 . 8% respectively ( Figure 2 ) . There was a decline in infection from baseline to 36 months in both the usual target group and the enhanced coverage group . At 36 months ( one year after the third MDA ) , the prevalence of infection was 4 . 0% in the usual target group and 5 . 4% in the enhanced target group , with an adjusted difference of 1 . 4% ( 95% Confidence Interval ( CI ) = −1 . 0% to 3 . 8% ) . The prevalence of follicular trachoma by treatment group showed a similar patter , with no difference between groups at baseline , 30 . 4% and 30 . 9% in the usual target and enhanced target group respectively ( Figure 3 ) . By 36 months , the trachoma prevalence had fallen to 6 . 1% and 9 . 0% respectively , an adjusted difference of 2 . 6% with 95% CI = − . 3% to 5 . 3% . Separate models of infection and trachoma at 36 months , adjusted for baseline prevalence , showed no effect of being in the enhanced target arm ( Table 3 ) . The prevalence of trachoma at baseline predicted the prevalence of trachoma at 36 months , but the prevalence of infection at baseline was not predictive of infection at 36 months . When analyzed as actual coverage , there was still no evidence that increasing coverage in children resulted in lower infection or trachoma prevalence at 36 months ( Table 4 ) . If the variable of average percentage coverage over the three rounds was replaced with a variable of percentage coverage at the last round of mass treatment , there was still no evidence for a significant decrease in infection or trachoma with increasing coverage at 36 months ( data not shown ) . There were no serious adverse events reported in either arm .
This community based , randomized trial had two major findings: first , there was no evidence of benefit to increasing mass drug administration coverage of children ages under ten years above 90% , compared to targeting coverage between 80–90% . The analyses repeated on actual coverage found no benefit per unit increase in coverage either . Second , that if the baseline prevalence of trachoma in communities are estimated at 20% or greater , two years of annual MDA were insufficient to decrease infection in any community below an estimate of 5% and at least three annual rounds will be necessary even with high coverage . There may be at least two possible reasons for the first finding . With coverage this high in children , there were very few children who were not treated at least once after three MDAs . In a study in these communities of persistent non- participation in MDA , Ssemanda et al found that only 2% of households contained children who did not participate in 2 MDAs [13] . By the third round of MDA , there may be little benefit to the extra effort required to achieve over 90% coverage . This supposition has supporting evidence with the rapid fall in infection by 36 months in both arms of the trial . Trachoma at baseline in these communities averaged 30% , with 22% infection . By three years and three rounds of MDA , infection averaged 4 . 7% and trachoma fell to an average of 7 . 6% . The trajectory suggests that just a few more rounds of MDA would be needed to decrease trachoma below 5% and even eliminate infection , fewer annual rounds than suggested by other work in Tanzania where coverage was estimated at less than 75% [6] . Another possible reason is that the difference in coverage in the enhanced target versus the usual target groups was relative small . The coverage in children in the usual target group was high , 90% , 88% , and 87% respectively for the three rounds . While some communities were lower , no community had MDA coverage in children below 80% , by design . The enhanced target group had significantly higher coverage , but the largest difference was in the third round with average coverage of 93% compared to the usual care group with average coverage of 87% . However , even adjusting for slightly higher rates of infection in the enhanced coverage group , there was no survey where the enhanced coverage group had a point estimate that was lower infection or trachoma compared to the usual coverage group . When the analyses were repeated using actual coverage in the communities , where the range was greater , there was no apparent effect of increasing coverage on infection or trachoma . The data appear to be more consistent with an absence of an effect of enhanced coverage than failure to detect a difference . It is unlikely that the intervention was carried out improperly in either arm of the trial . Treatment verification consistently showed high agreement between recorded treatment and personal history of recipients . Trachoma and infection fell over time consistent with good coverage in children as reported from studies elsewhere in other high prevalence villages [14] , [15] . The fact that we did not stop MDA in any of the communities randomized to the cessation rule , even with high coverage , provides support for the WHO guideline that suggest 3 rounds of MDA before re-assessing the impact , at least for communities with baseline prevalence above 20% . We had expected that with high coverage , communities would reach our pre-specified target of zero cases of infection in 100 sentinel children before the third round of MDA . This did not happen when the starting prevalences of infection in the communities were , on average , 22% . Our experience was different than the experience with high coverage in small communities in Ethiopia , where after a single round of high coverage , infection fell from 50% to less than 5% at 6 months and after a second round at 6 months , disappeared [15] . Other studies in The Gambia and in a low prevalence community in Tanzania reported that one to two annual rounds was sufficient to eliminate infection [7] , [16] , but these communities started with low levels of infection . A recent study from Ethiopia noted that annual treatment with high coverage reduced infection from 42% in children ages 0–9 years to 1 . 9% after four rounds of annual MDA , which is more consistent with the trajectory of our observed decline [17] . In that study , 5 of the 12 communities had no infection at 36 months , but with on average only 50 sentinel children per community , the possibility of infection as high as 5% could not be ruled out . In summary , a community randomized clinical trial comparing high coverage ( 80–90% ) in children with antibiotic for trachoma to very high coverage ( >90% ) found no difference in infection or trachoma rates after three annual rounds of MDA . The results suggest that aiming for at least 80% coverage of children in trachoma endemic communities is reasonable , and there is no advantage to expending resources to push for even higher coverage . When treating communities with 20% or greater prevalence of trachoma at baseline , at least 3 rounds of MDA will be needed before infection drops confidently below 5% . | The World Health Organization recommends at least 3 annual antibiotic mass drug administrations ( MDA ) where the prevalence of trachoma is >10% in children ages 1–9 years , with coverage at least at 80% . However , the additional value of higher coverage targeted at children with multiple rounds is unknown . We randomized 32 communities in Kongwa , Tanzania , with starting prevalence estimated at >20% to four arms: annual MDA aiming for coverage of children between 80%–90% ( usual target ) versus aiming for coverage>90% ( enhanced target ) ; and to: MDA for three years versus a rule of cessation of MDA early if the estimated prevalence of ocular C . trachomatis infection was less than 5% . After three rounds of MDA , infection with C . trachomatis and trachoma had declined significantly from baseline but no communities had treatment stopped . There was no difference in infection or in trachoma at three years comparing the usual coverage communities to the enhanced coverage communities . We conclude that in communities that had pre-treatment prevalence of follicular trachoma of 20% or greater , there is no evidence that MDA can be stopped before 3 annual rounds , even with high coverage . Increasing coverage in children above 90% does not appear to confer additional benefit . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [
"medicine",
"neglected",
"tropical",
"diseases",
"infectious",
"diseases",
"trachoma"
] | 2013 | A Randomized Trial of Two Coverage Targets for Mass Treatment with Azithromycin for Trachoma |
With the opportunistic pathogen Pseudomonas aeruginosa , quorum sensing based on homoserine lactones was found to influence biofilm formation . Here we discern a mechanism by which quorum sensing controls biofilm formation by screening 5850 transposon mutants of P . aeruginosa PA14 for altered biofilm formation . This screen identified the PA3885 mutant , which had 147-fold more biofilm than the wild-type strain . Loss of PA3885 decreased swimming , abolished swarming , and increased attachment , although this did not affect production of rhamnolipids . The PA3885 mutant also had a wrinkly colony phenotype , formed pronounced pellicles , had substantially more aggregation , and had 28-fold more exopolysaccharide production . Expression of PA3885 in trans reduced biofilm formation and abolished aggregation . Whole transcriptome analysis showed that loss of PA3885 activated expression of the pel locus , an operon that encodes for the synthesis of extracellular matrix polysaccharide . Genetic screening identified that loss of PelABDEG and the PA1120 protein ( which contains a GGDEF-motif ) suppressed the phenotypes of the PA3885 mutant , suggesting that the function of the PA3885 protein is to regulate 3 , 5-cyclic diguanylic acid ( c-di-GMP ) concentrations as a phosphatase since c-di-GMP enhances biofilm formation by activating PelD , and c-di-GMP inhibits swarming . Loss of PA3885 protein increased cellular c-di-GMP concentrations; hence , PA3885 protein is a negative regulator of c-di-GMP production . Purified PA3885 protein has phosphatase activity against phosphotyrosine peptides and is translocated to the periplasm . Las-mediated quorum sensing positively regulates expression of the PA3885 gene . These results show that the PA3885 protein responds to AHL signals and likely dephosphorylates PA1120 , which leads to reduced c-di-GMP production . This inhibits matrix exopolysaccharide formation , which leads to reduced biofilm formation; hence , we provide a mechanism for quorum sensing control of biofilm formation through the pel locus and suggest PA3885 should be named TpbA for tyrosine phosphatase related to biofilm formation and PA1120 should be TpbB .
Pseudomonas aeruginosa , an opportunistic pathogen , is often used to elucidate how biofilms form because persistence of this bacterium is linked to its ability to form biofilms [1] . Biofilms are formed by the attachment of bacteria to submerged surfaces in aquatic environments through their production of microbial products including polysaccharides , proteins , and nucleic acids [1] . In P . aeruginosa PA14 , the glucose-rich extracellular polysaccharide ( EPS ) of the biofilm matrix is formed by proteins encoded by the pel operon; note the related strain P . aeruginosa PAO1 has two EPS production loci , pel and psl [2] , [3] . Mutations in the pel locus of P . aeruginosa PA14 dramatically decrease biofilm formation as well as pellicle formation; pellicles are formed at the interface between the air and liquid medium [3] . Regulation of Pel polysaccharide involves 3 , 5-cyclic diguanylic acid ( c-di-GMP ) which is formed by diguanylate cyclases with GGDEF motifs that synthesize this second messenger; phosphodiesterases with EAL motifs catabolize c-di-GMP . Many proteins with GGDEF motifs enhance biofilm formation [4]; for example , c-di-GMP increases cellulose biosynthesis in Acetobacter xylinus [5] , and c-di-GMP enhances EPS production by binding the PelD protein that is a c-di-GMP receptor in P . aeruginosa PA14 [6] . Thus , biofilm formation is controlled by a signal cascade mediated by a complex of c-di-GMP and PelD in P . aeruginosa PA14; however , the upstream portions of this cascade have not been elucidated [7] . Quorum sensing ( QS ) is bacterial communication using diffusible molecules known as autoinducers to regulate population behavior and is related to both polysaccharide production and biofilm formation . To date , three QS systems have been identified in P . aeruginosa . Las-based QS is regulated by N- ( 3-oxododecanoyl ) -L-homoserine lactone , produced by LasRI [8] , and Rhl-based QS is regulated by N-butyryl homoserine lactone , produced by RhlRI [9] . The third QS molecule , 2-heptyl-3-hydroxy-4-quinolone ( PQS ) , was identified as a regulator for both Las- and Rhl-QS [10] . These cell communication signals regulate several phenotypes including virulence and antibiotic resistance [11] . Although the relationship between QS and biofilm formation has not been fully elucidated , some lines of evidence show the importance of QS for biofilm formation . Cells lacking Las-QS in P . aeruginosa form flat biofilms , and this structural abnormality makes bacteria in biofilms more sensitive to antibiotic treatment [12] . Biofilm architecture is regulated by rhamnolipids whose synthesis is controlled by Rhl-QS [13] . Thus , P . aeruginosa QS seems to participate in the development of biofilm architecture rather than initiation of biofilm formation . In addition , LasR- and RhlR-QS have been shown to influence the pel operons indirectly and another transcriptional regulator that controls pel has been predicted [7] . Protein phosphorylation and dephosphorylation are well-conserved posttranslational modifications in both prokaryotes and eukaryotes [14] . Protein kinases and phosphatases modulate cellular activity by adding and removing phosphate groups at Ser , Thr , or Tyr residues . Phosphorylation also occurs at His and Asp residues by histidine kinases and response regulators in two-component regulatory systems . Although the discovery of protein phosphorylation was delayed in prokaryotes compared to eukaryotes [14] , many genome sequences predict the existence of phosphorylation/dephosphorylation systems in prokaryotes . The P . aeruginosa genome encodes an extraordinary number of the genes for two-component regulatory systems [15] , and diverse cellular functions are regulated by His-Asp phosphorylation including chemotaxis , iron acquisition , alginate production , and virulence factors [16] . In contrast , phosphorylation at Ser , Thr , and Tyr residues has not been studied well in P . aeruginosa; although , Fha1 of the Type VI secretion system is posttranslationally regulated through Thr phosphorylation by the protein kinase PpkA and dephosphorylated by the phosphatase PppA [17] . In Bacillus subtilis , mutations in prkC , a Ser/Thr kinase , and prpC , a phosphatase , decrease sporulation and biofilm formation [18] . Mutations in stk1 , a Ser/Thr kinase , and stp1 , a phosphatase of Stk1 , decrease virulence in Streptococcus agalactiae [19] . These findings show that posttranslational modification via protein phosphorylation at Ser , Thr , and Tyr residues regulates various cellular functions . In this study , our goal was to explore the complex regulatory cascade that includes detection of QS signals , Pel polysaccharide production , and biofilm formation . By screening 5850 transposon mutants for altered biofilm formation , we identified and characterized a novel protein tyrosine phosphatase , TpbA ( tyrosine phosphatase related to biofilm formation ) , that represses biofilm formation through the pel locus . The tpbA mutant displays pleiotropic phenotypes such as hyperbiofilm formation , enhanced EPS production , altered colony morphology , increased aggregation , elevated c-di-GMP , and abolished swarming . Loss of an uncharacterized GGDEF protein , PA1120 ( TpbB ) , suppressed these phenotypes , indicating that TpbA controls c-di-GMP production through TpbB . Therefore , the mechanism for QS-control of biofilm formation has been extended to include a novel phosphatase ( TpbA ) , a diguanylate cyclase ( TpbB ) , and c-di-GMP; hence , the predicted additional level of control of the pel polysaccharide locus has been identified and involves c-di-GMP as controlled by a tyrosine phosphatase .
Previously , by screening 5850 transposon mutants for altered biofilm formation , we identified 137 transposon mutants of P . aeruginosa PA14 with over 3-fold enhanced biofilm formation [20] . Among these mutants , the tpbA ( PA3885 ) mutant increased biofilm formation by 147-fold after 8 h in LB medium at 37°C ( Fig . 1A ) . This significant increase in biofilm formation upon inactivating tpbA is partially due to enhanced attachment to the polystyrene surface because biofilm formation at the bottom of the plates ( solid/liquid interface ) increased gradually with the tpbA mutant while PA14 did not form biofilm on the bottom of the plate ( Fig . 1B ) . Motility often influences biofilm formation in P . aeruginosa; biofilm formation is inversely influenced by swarming motility [21] , and swimming motility increases initial attachment to surfaces during biofilm development [22] . To examine the relationship between enhanced biofilm formation and motility in the tpbA mutant , we examined swimming and swarming motility for this mutant; rhlR [23] and flgK [22] mutants were used as negative controls for swarming and swimming motility , respectively . Although PA14 swarmed on the surface of plates at 24 h , the tpbA mutations abolished swarming like the rhlR mutation ( Fig . 2A ) . The tpbA mutation also decreased swimming motility by 40% ( Fig . 2B ) . Swarming is positively influenced by production of the biosurfactant putisolvins in P . putida [24] and rhamnolipids in P . aeruginosa [23] . However , no significant difference was found in the production of rhamnolipids between PA14 and the tpbA mutant ( Fig . 2C ) . This shows the tpbA mutation abolishes swarming in a manner distinct from the production of rhamnolipids . Congo-red is often used to observe colony morphology because it detects EPS production and this impacts biofilm formation; for example , the wspF mutant shows wrinkly colony morphology on Congo-red plates and has increased biofilm formation [25] , while smooth colonies like the pelA mutant [3] form less biofilm . We found that the tpbA mutant formed a red , wrinkly colony when it was grown on Congo-red plates at 37°C , although PA14 and the pelA mutant formed white smooth colonies ( Fig . 3A ) . When the bacteria were grown at 25°C , both PA14 and the tpbA mutant formed red wrinkly colonies , but the pelA mutant still formed a white smooth colony ( Fig . 3A ) . These observations with the pelA mutant and wild-type PA14 are identical to the previous report that expression of the pel genes is induced at room temperature and repressed at 37°C [7] . Therefore , the red wrinkly colony formed by the tpbA mutant at 37°C implies increased production of EPS via Pel . We also quantified the amount of EPS bound to cells of PA14 and the tpbA mutant at both 37°C and 25°C . As shown in Fig . 3B , the tpbA mutant produced 28-fold more EPS than PA14 at 37°C . The tpbA mutant also produced 4 . 3-fold more EPS than PA14 at room temperature . The pelA mutant ( negative control ) did not form EPS at both temperatures tested . We also found that the tpbA mutant formed a pronounced pellicle at 37°C after 1 day , but PA14 and the pelA mutant did not form a pellicle ( data not shown ) . At 25°C , both the tpbA mutant and PA14 formed pellicles after 5 days . Taken together with the EPS production data , TpbA reduces pellicle formation by decreasing Pel activity . To confirm the impact of the tpbA mutation on pel expression and to investigate its impact on the whole genome , a whole-transcriptome analysis was performed with biofilm cells of the tpbA mutant at 37°C at 7 h; planktonic cells were not assayed since we were primarily interested in how TpbA controls biofilm formation . Inactivation of tpbA altered diverse loci including genes related to EPS production ( pelACDF induced approximately 4-fold ) , transport ( PA2204 repressed approximately 5-fold , PA4142–PA4143 induced approximately 3-fold ) , type IV pili ( PA4302 to PA4306 repressed approximately 4-fold ) , and a putative adhesin and its regulator ( PA4624–PA4625 induced approximately 4-fold ) ( Tables S1 and S2 ) . Expression of tpbA was induced as much as 120-fold in the tpbA mutant , suggesting that TpbA negatively regulates its transcription . The whole-transcriptome experiments were performed twice using independent cultures of PA14 and the tpbA mutant at 7 h , and most of the differentially regulated genes were consistently altered except pel genes which were induced the most in the samples containing an RNase inhibitor . A whole-transcriptome analysis was also conducted using biofilm cells at 4 h since the mode of growth switched from planktonic to biofilm for the tpbA mutant at this time ( Fig . 1A ) . Similar to the 7 h results , several loci were induced including pelAEF ( 1 . 5- to 1 . 7-fold ) , tpbA ( 42-fold ) , PA1168–PA1169 ( 1 . 4- to 2 . 1-fold ) , PA3886 ( 3 . 5-fold ) , and PA4624–PA4625 ( 2- to 3 . 7-fold ) ( Table S1 ) . To verify induction of the pel locus , expression of pelA was determined by quantitative real time-PCR ( qRT-PCR ) . Using two independent RNA samples extracted from biofilm cells at 7 h , pelA was induced 112±100-fold in the tpbA mutant vs . PA14 . These results showed EPS production is induced significantly in the tpbA mutant due to overexpression of pel genes . qRT-PCR also confirmed induction of PA4625 ( 7±7-fold ) as well as PA4139 ( 38±30-fold ) that encodes a hypothetical protein . Cell aggregative behavior is also related to biofilm formation so we investigated the role of TpbA on cell aggregation and found the tpbA mutant causes cell aggregation ( Fig . 4A ) . Autoaggregation of the tpbA mutant was also observed in 96-well polystyrene plates during biofilm formation ( data not shown ) . Our whole-transcriptome analysis showed that inactivating tpbA induced both PA4624 ( encodes for a putative hemolysin activator ) and PA4625 ( encodes for an adhesin/hemagglutinin ) by 2 . 1- to 4 . 9-fold . In E . coli , adhesin regulates cell aggregation as well as attachment [26] . To examine whether PA4624–PA4625 control adhesive activity in P . aeruginosa , we investigated biofilm formation with these mutants . Both mutants showed decreased initial biofilm formation; i . e . , initial attachment , to polystyrene plates at 1 h and 2 h ( Fig . 4B ) , and final biofilm formation at 24 h was also decreased for both the PA4624 and PA4625 mutants , which suggests that both gene products control attachment to the surface . Therefore , TpbA decreases cell aggregation probably by repressing the PA4624 and PA4625 genes . To verify whether the phenotypes observed in the tpbA mutant were caused by loss of function of TpbA , we confirmed transposon insertion in tpbA by PCR at residue 25 . Furthermore , biofilm formation for both PA14 and the tpbA mutant were examined with tpbA expressed in trans under the control of an arabinose-inducible promoter . tpbA expression reduced biofilm formation of the tpbA mutant by 33% ( Fig . S1A ) and abolished biofilm formation on the bottom of the plates ( Fig . S1B ) . Similar results were found upon expressing tpbA in wild-type PA14 ( OD540 value was 0 . 22±0 . 02 for PA14/pMQ70 and 0 . 02±0 . 01 for PA14/pMQ70-tpbA , Fig . S1A ) . In addition , the aggregative phenotype of the tpbA mutant was also complemented by expression of tpbA in trans ( Fig . S1C ) . Taken together , TpbA functions as a negative regulator of biofilm formation and aggregation in PA14 . To investigate how TpbA regulates biofilm formation , EPS production , wrinkly colony morphology , and cell aggregation , genetic screening was conducted using Tn5-luxAB transposon mutagenesis to find suppressive loci for the phenotypes of the tpbA mutation . The double mutant library ( tpbA plus random gene inactivations ) was screened first for a reduction in aggregation; this step eliminated most cells with unaltered phenotypes by allowing them to aggregate and precipitate at the bottom of the tube . The cells remaining in the supernatant that failed to aggregate like the tpbA mutant were grown on Congo-red plates , incubated at 37°C for 3–4 days , and colonies displaying a white and smooth shape like the wild-type strain were chosen . After that , a third screen was performed by assaying biofilm formation using 96-well polystyrene plates to identify double mutants that had biofilm formation like the wild-type strain . Twenty-six mutants were identified that showed reduced aggregation , a white smooth colony , and reduced biofilm formation like the wild-type strain , and 19 of these mutations were in the pel locus ( Fig . 5A , Table 1 ) . Four of the other mutants have the Tn5-luxAB insertion in the TpbB gene ( encodes a GGDEF-motif protein ) and in the PA1121 gene ( encodes a hypothetical protein ) . In addition , insertions were found in the PA1678 gene ( encodes a putative DNA methylase ) and in the promoter of the PA5132 gene ( encodes a putative protease ) ( Table 1 ) . Like the double mutants , all of the single mutants lacking the gene identified by genetic screening were tested for biofilm formation , and all of these mutants formed less biofilm as reported previously ( Fig . 5 ) [3] , [4] . Results of genetic screening and the whole-transcriptome analysis implied TpbA regulates c-di-GMP concentrations since loss of one of the GGDEF proteins ( TpbB ) masked the phenotypes of the tpbA mutant . tpbB encodes a functional GGDEF protein whose activity was confirmed by overexpressing this gene in P . aeruginosa [4] . We also confirmed that expression of tpbB increases cell aggregation and attachment to tubes so the tpbB mutation may be complemented ( Fig . S2 ) . In addition , we measured the cellular c-di-GMP concentrations of PA14 and the tpbA mutant using high performance liquid chromatography ( HPLC ) as reported previously [4] . The peaks corresponding to c-di-GMP were observed with the extracts of the tpbA mutant , but not with those of PA14 , and the peak was confirmed by comparing the spectrum to purified c-di-GMP as well as by spiking the samples with purified c-di-GMP ( Fig . S3 ) . We estimated the cellular c-di-GMP concentration was 10±2 pmol/mg cells in the tpbA mutant . This is comparable to the c-di-GMP concentration found for a small colony variant that showed aggregation ( around 2 . 0 pmol/mg cells ) and a mutant with wrinkly colony morphology [27] . Overexpression of tpbB results in c-di-GMP concentrations of 134 pmol/mg cells in PA14 [4] . Therefore , TpbA reduces c-di-GMP concentrations in the cell and probably does so via TpbB . tpbA encodes a 218 aa protein that has the conserved domain for a protein tyrosine phosphatase [28] , [29] since it has the C ( X ) 5R ( S/T ) motif beginning at aa 132 ( CKHGNNRT ) . To confirm it is a tyrosine phosphatase , we purified TpbA by adding a polyhistidine tag at either the N-terminus ( TpbA-nHis ) or the C-terminus ( TpbA-cHis ) ( note only the C-terminus fusion protein was active ) . Expression of recombinant TpbA was confirmed in E . coli by clear expression of a band at 24 kD ( Fig . 6A ) . The purified TpbA protein had phosphatase activity with p-nitrophenyl phosphate ( pNPP ) that is often used as a general phosphatase substrate [29] ( Fig . 6B ) . Further proof that TpbA is a tyrosine phosphatase was found using a tyrosine phosphatase specific inhibitor , trisodium orthovanadate [30] , that completely inhibited the phosphatase activity of TpbA-cHis ( Fig . 6B ) . The third and fourth lines of evidence that TpbA is a tyrosine phosphatase were found using tyrosine specific substrates; TpbA-cHis dephosphorylated both phosphotyrosine peptides , END ( pY ) INASL ( peptide type I ) and DADE ( pY ) LIPQQG ( peptide type II ) ( Fig . 6C ) , and this activity was inhibited by trisodium orthovanadate . These results show conclusively that TpbA encodes a tyrosine phosphatase . To see the effect of tyrosine phosphorylation on biofilm formation , biofilm formation was examined in PA14 with trisodium orthovanadate at 37°C for 4 h which should reduce dephosphorylation by TpbA . Trisodium orthovanadate increased PA14 biofilm formation 3 . 6-fold ( Fig . S4 ) , showing that cellular tyrosine phosphorylation increases biofilm formation . The N-terminal region of TpbA protein has a putative signal peptide , predicted by pSORT [31] , that appears necessary for secretion of this protein ( 28 aa , MHRSPLAWLRLLLAAVLGAFLLGGPLHA ) . This implied that processing of N-terminal region of TpbA protein may be essential for full phosphatase activity . To prove that TpbA has an active signal sequence , we expressed TpbA in E . coli and collected the proteins from cytosolic , periplasmic , and membrane fractions . All fractioned proteins were analyzed by SDS-PAGE , and we found that TpbA exclusively localized in the periplasm ( data not shown ) . Hence , TpbA probably dephosphorylates its substrate in the periplasm which explains why phosphatase activity was seen only with the fusion protein with the His tag at the C-terminus . Since TpbA is a tyrosine phosphatase that is found in the periplasm and since TpbB has three likely periplasmic tyrosines ( Y48 , Y62 , and Y95 ) [32] , we mutated the periplasmic tyrosine residues by converting them to phenylalanine and checked for TpbB activity in the tpbB mutant . Active TpbB , from overexpression of tpbB using the tpbB mutant and pMQ70-tpbB , always leads to aggregation whereas the empty plasmid does not cause aggregation ( Fig . S5 ) ; hence , if a necessary tyrosine is mutated , there should be a reduction in aggregation . Aggregation was always observed with TpbB-Y95F in nine cultures; hence TpbB-Y95F remains active even though it lacks tyrosine 95 so this tyrosine is not phosphorylated/dephosphorylated . In contrast , the Y48F mutation of TpbB decreased aggregation for 43% of the cultures ( 20 of 46 cultures did not aggregate ) , and the Y62F mutation decreased aggregation for 24% of the cultures ( 9 of 37 cultures did not aggregation ) . Hence , both Y48 and Y62 are likely targets for tyrosine phosphorylation/dephosphorylation of TpbB with Y48 preferred . We confirmed that these mutations did not affect expression level of TpbB protein ( data not shown ) . Tyrosine phosphorylation and dephosphorylation have crucial roles in cellular signaling and are well-conserved among many organisms [33] . Some bacterial tyrosine phosphorylations have been identified and these regulatory systems control divergent cellular functions [34] . In order to predict whether TpbA function is conserved among other species , we conducted a BLASTP search and found the TpbA protein is highly conserved among P . aeruginosa ( PAO1 , PA14 , C3719 , and PA7 with an E value less than 3e-98 ) and is well-conserved among P . fluorescens Pf-5 , P . fluorescens Pf0-1 , P . mendocina , Burkholderia cepacia , Pelobacter carbinolicus , Desulfatibacillum alkenivorans , Bacteroides thetaiotaomicron , B . ovatus , B . caccae , Acinetobacter baumannii , Desulfococcus oleovorans , and Geobacter metallireducens ( E values less than 3e-11 ) . All of these conserved tyrosine phosphatases have the C ( X ) 5R ( S/T ) signature and most are uncharacterized . Even though protein similarity is not very high , some eukaryotes , such as Homo sapiens and Arabidopsis thaliana , have TpbA homologs with a C ( X ) 5R ( S/T ) signature . Therefore , TpbA and TpbA homologs may share important functions in procaryotes and eucaryotes . QS regulates many genes in P . aeruginosa via a conserved cis-element in the promoter of each gene . N- ( 3-oxododecanoyl ) -L-homoserine lactone binds to the LasR transcriptional regulator [35] , and this complex interacts with the las-box , defined as CT- ( N ) 12-AG sequence [36] . The Las-box is conserved among the promoters of the Las-QS regulated genes including lasB , rhlAB , and rhlI [36] . Another class of transcriptional regulation is governed by the lys-box , that is defined as a palindromic sequence , T- ( N ) 11-A [37] , and MvfR is a LysR-type transcription factor that binds to the lys-box [38] . We found that the promoter of tpbA ( ptpbA ) has a putative las-box 220 bp upstream of the start codon ( CTCGCCTCGCTGAAAG ) and a putative lys-box 90 bp upstream of the start codon ( TGAAGCTGCCTCA ) . In order to examine if expression of tpbA is regulated by QS , we constructed a ptpbA::lacZ fusion plasmid ( pLP-ptpbA ) and transformed this into QS-related PA14 mutants ( lasI , rhlI , and lasR rhlR ) . Expression of tpbA gene in biofilm cells was reduced by 42% in the lasI mutant , but not in the rhlI mutant ( Fig . 7 ) . Corroborating these results , inactivation of both lasR and rhlR also decreased expression of tpbA gene by 39% ( Fig . 7 ) . Similar results were obtained when the activity of ptpbA::lacZ was examined in planktonic cells ( 50% reduction in transcription for the lasI mutant and 37% reduction for the lasR rhlR mutant ) . Since loss of QS only affected expression of tpbA by 50% , other factors may also participate in the regulation of tpbA . These results suggest that Las-QS , rather than Rhl-QS , is an activator of tpbA expression with other unknown regulators . We also investigated whether the tpbA mutation influences the regulation of Las- and Rhl-QS using plasR::lacZ and prhlR::lacZ plasmids . Expression of lasR was slightly increased ( 1 . 3-fold ) with the tpbA mutation . This indicated that Las-QS has more impact on expression of tpbA than tpbA does on that of Las-QS . In addition , expression of rhlR was decreased by 2-fold in the tpbA mutant . Hence , LasR appears to enhance tpbA transcription and TpbA leads to increased rhl transcription .
In this study , we demonstrate that TpbA is a tyrosine phosphatase that regulates diverse phenotypes in P . aeruginosa including the concentration of cellular c-di-GMP . As a second messenger , c-di-GMP is a positive regulator of biofilm formation [4] , EPS production [6] , and pellicle formation [4] , and a negative regulator of swarming motility [39] . The lines of evidence that show TpbA represses c-di-GMP production in P . aeruginosa that we found are ( i ) inactivating tpbA increases c-di-GMP ( Fig . S3 ) ; ( ii ) inactivating tpbA increases biofilm formation ( Fig . 1 ) , EPS production ( Fig . 3B ) , and pellicle formation , and c-di-GMP stimulates biofilm formation [4] , EPS production [6] , and pellicle formation [4]; ( iii ) inactivating tpbA increases expression of the pel locus ( seen via the whole-transcriptome analysis and RT-PCR ) , and c-di-GMP activates expression of pelA [6]; ( iv ) inactivating tpbA increases aggregation ( Fig . 4 ) and expression of adhesins ( PA4625 ) , and c-di-GMP stimulates adhesion [40]; ( v ) inactivating tpbA decreases motility ( abolishing swarming and decreasing swimming in the tpbA mutant , Fig . 2AB ) , and c-di-GMP decreases swarming [40]; ( vi ) inactivating tpbB ( encodes a GGDEF-motif protein that produces c-di-GMP [4] ) suppresses the phenotypes observed in the tpbA mutant , and ( vii ) expression of tpbA and tpbB in trans complements aggregation/biofilm formation and aggregation , respectively . Thus , TpbA represses these phenotypes by decreasing c-di-GMP . A proposed regulatory mechanism for biofilm formation by TpbA is shown in Fig . 8 . EPS production in P . aeruginosa PA14 is regulated by PelA [3] . Transcription of pelA is higher at temperatures lower than 37°C , and PA14 forms more biofilm at lower temperatures [7] . However , the tpbA mutation seems to constitutively enhance pel expression independently from this temperature regulation as seen in the enhanced EPS production at 37°C ( Fig . 3 ) and the whole-transcriptome analysis that was conducted at 37°C ( Table S2 ) . In addition to increased expression of pelA , additional activation of Pel proteins might be caused by the increased c-di-GMP concentration by the tpbA mutation since c-di-GMP binds PelD and increases EPS production [6] . In addition to the enhanced EPS production ( Fig . 3B ) and increased pel expression ( Fig . 3A , Table S1 , and qRT-PCR ) seen in the tpbA mutant , another reason why inactivating tpbA increased biofilm formation is the elevated adhesin activity as seen via enhanced biofilm formation on the bottom of polystyrene plates ( Fig . 1B ) . Cell surface adhesins affect bacterial adhesive activity [41] , and we have discovered a novel adhesin ( PA4625 ) that is related to TpbA ( Table S1 ) and to initial biofilm formation ( Fig . 4B ) . Since expression of adhesion factors is also positively regulated by c-di-GMP [40] , elevated c-di-GMP level enhances adhesion of the tpbA mutant . c-di-GMP seems to control the switch of motility-sessility of the tpbA mutant since inactivation of TpbA abolished swarming motility ( Fig . 2A ) and decreased swimming motility by 40% ( Fig . 2B ) , although regulation of swarming motility is very complex as its activity is controlled by QS , flagellar synthesis , and production of rhamnolipids [42] . In addition , our whole-transcriptome results showed weak repression of some of flagellar biosynthesis genes ( flg , fle , and fli loci ) due to the elevated c-di-GMP , and activity of FleQ , a transcriptional activator of flagellar biosynthesis , is repressed upon binding c-di-GMP [43] . Hence , the increased c-di-GMP concentrations may repress motility of the tpbA mutant via the FleQ pathway that affects expression of flagellar synthesis genes . Many genes are expected to be differentially regulated by changing c-di-GMP concentrations since it plays a role as a second messenger in P . aeruginosa . Similar regulation of gene expression was observed between the tpbA mutant and the other strains related to c-di-GMP production . For example , production of PA1107 , TpbB , and PA3702 proteins that have a GGDEF-domain leads to activation of pelA expression [6] . Mutation in wspF , encoding a CheB-like methylesterase , increases both biofilm formation and c-di-GMP production [25] . wspF mutation altered expression of genes such as pelABCDEFG , PA4624 , PA4625 , PA2440 , and PA2441 whose expression are induced in the tpbA mutant ( Table S2 ) . Common regulation of these genes may be partially controlled by the elevated cellular c-di-GMP concentrations . In contrast , expression of pelA was not induced in the bifA mutant that produces more c-di-GMP [44] . This may be because regulation of c-di-GMP signaling is complex in that the P . aeruginosa genome encodes 17 diguanylate cyclases , 5 phosphodiesterases , and 16 diguanylate cyclase-phosphodiesterase proteins [4] . Relevance of c-di-GMP to regulation of diverse cellular functions is now an emerging topic in bacteriology . This second messenger is an activator of cellulose synthase in Acetobactor xylinum [5] and controls many phenotypes in P . aeruginosa [4] . Several GGDEF proteins for synthesis and EAL proteins for degradation of c-di-GMP have been identified in P . aeruginosa [6] , [39] , [44] , and increased production of c-di-GMP enhances biofilm formation and decreases swarming motility [4] , [6] , [39] , [44] . Similarly , enhanced biofilm formation and/or abolished swarming motility were observed in the tpbA mutant via increased production of cellular c-di-GMP . Since TpbA does not possess GGDEF and EAL domains , this protein indirectly influences cellular c-di-GMP concentrations via its phosphatase activity as shown by activity with both pNPP , a broad substrate for phosphatases , and two phosphotyrosine-specific peptides ( Fig . 6 ) . We also found that processing the N-terminal signal sequence may be necessary for TpbA activity in the periplasm . Hence , our results reveal a novel regulatory mechanism for cellular c-di-GMP concentration by tyrosine phosphorylation in the periplasm of P . aeruginosa; control of c-di-GMP by tyrosine phosphorylation has not been shown previously . There is little known about the regulation of GGDEF and EAL proteins in regard to regulation of c-di-GMP level . A chemosensory system , encoded by wspABCDEFR in PAO1 , regulates c-di-GMP production via a His-Asp phosphorylation relay [25] . For the tpbA mutant , tpbB was found to reverse the phenotype of tpbA , suggesting that overproduction of c-di-GMP is clearly related to the phenotypes of the tpbA mutant as overexpression of this gene caused pronounced aggregation ( Fig . S2 ) . Probably , TpbB , or another GGDEF protein , might participate in c-di-GMP synthesis in the tpbA mutant . A comprehensive analysis of all of the P . aeruginosa GGDEF proteins has been completed , and those GGDEF proteins that abolished or decreased biofilm formation are PA0169 , PA1107 , TpbB , PA1181 , PA1433 , PA1727 , PA3702 , PA4959 , and PA5487 [4] . Among these GGDEF proteins , only PA1107 , TpbB , PA3702 , and PA5487 increased biofilm formation when their genes were overexpressed [4] . Because TpbA is a periplasmic protein , its target GGDEF protein should have periplasmic regions . By a bioinformatics evaluation , of those four GGDEF proteins that increased biofilm formation , only PA1107 and TpbB have transmembrane regions . Taken together with the results of genetic screening , TpbB is the most likely target protein for TpbA . Also , our results imply the periplasmic Y48 and Y62 residues of TpbB are the likely targets for tyrosine phosphorylation . We are now investigating whether TpbA regulates the activity of GGDEF proteins to control cellular c-di-GMP concentrations in P . aeruginosa . The relationship between tyrosine phosphorylation and biofilm formation is not well established . We found that trisodium orthovanadate treatment increased biofilm formation of PA14 ( Fig . S4 ) , indicating that tyrosine phosphorylation increases biofilm formation in P . aeruginosa . Recently , Ltp1 , a low molecular weight tyrosine phosphatase in non-motile , Gram-negative P . gingivalis , was identified as a negative regulator of EPS production and biofilm formation [45] . A sequence similarity search shows TpbA is not a homolog of Ltp1 , because TpbA has a signal sequence in its N-terminal region , and TpbA is translocated into the periplasm . Other differences were found in the position of the motif for the tyrosine phosphatase , since TpbA has the motif at the position 132 and Lpt1 has it at position 9 . It appears the P . aeruginosa genome encodes another tyrosine phosphatase , annotated as ptpA , that has a tyrosine phosphatase motif at the position 7 and does not have a signal sequence at the N-terminus . The function of PtpA is unknown but it is essential [46] . In contrast to poorly-investigated Tyr phosphorylation , regulation of biofilm formation by phosphorylation has been identified for several systems; for example , for the His kinase/Asp response regulator phosphorylation systems RocS1/RocA1/RocR of P . aeruginosa PAK [47] and the PAO1 PA1611/PA1976/PA2824/RetS/HptB system of P . aeruginosa [48] . In addition , in the B . subtilis PrkC/PrpC system [18] , loss of a membrane-anchored Thr kinase and its phosphatase reduces biofilm formation . Our results indicate that TpbA acts as a negative regulator of cellular c-di-GMP formation and loss of TpbA results in increased c-di-GMP concentrations that enhance biofilm formation and inhibit motility . These results show clearly that posttranslational modification through phosphatase activity is related to bacterial biofilm formation as well as to the regulation of the synthesis of cellular second messengers . In addition , by showing tpbA transcription is increased by LasR ( Fig . 7 ) and by finding AHL-binding motifs , we have now linked quorum sensing to c-di-GMP concentrations and biofilm formation in P . aeruginosa . Similarly , Vibrio cholerae QS was found recently to reduce cellular c-di-GMP concentrations via a c-di-GMP-specific phosphodiesterase which leads to lower biofilm formation [49] . A common element in both studies is that QS seems to be a negative regulator of c-di-GMP . The tpbA mutation caused a hyper-aggregative phenotype ( Fig . 4 ) , and this would lead to flat biofilms since the wspF mutant , which accumulates increased c-di-GMP , formed flat biofilms [25] . Formation of flat and undifferentiated biofilms is also observed by loss of LasI function [12] that can activate tpbA expression . Hence , TpbA might participate in developing biofilm structure . These results are important in that the regulatory networks that control c-di-GMP concentrations are now linked to the environment and cell populations .
Strains used in this study are listed in Table 2 . P . aeruginosa PA14 wild-type and its isogenic mutants were obtained from the Harvard Medical School [46] . Transposon insertion of the tpbA mutant was verified as described previously with a minor modification [50] . Briefly , the tpbA gene was amplified from chromosomal DNA using primers PA14_13660-VF and PA14_13660-VR ( Table S3 ) which did not amplify chromosomal DNA from the tpbA mutant . In addition , the DNA fragment corresponding to the end of the transposon and tpbA gene was amplified with tpbA chromosomal DNA using primers PA14_13660-VF and GB-3a ( Table S3 ) and PA14_13660-VR and R1 ( Table S3 ) but these pairs of primers did not amplify PA14 wild-type chromosomal DNA . P . aeruginosa and E . coli were routinely grown in Luria-Bertani ( LB ) medium at 37°C unless noted . Gentamicin ( 15 µg/mL ) and tetracycline ( 75 µg/mL ) were used for growth of the P . aeruginosa transposon mutants , carbenicillin ( 300 µg/mL ) was used to maintain P . aeruginosa plasmids , and kanamycin ( 50 µg/mL ) and chloramphenicol ( 50 µg/mL ) were used to maintain E . coli plasmids ( Table 2 ) . For complementation of the tpbA and tpbB mutations , tpbA and tpbB were expressed under the control of the pBAD promoter in pMQ70 [51] . tpbA and tpbB were amplified using a Pfu DNA polymerase with primers PA14_13660-F1-NheI and PA14_13660-R-cHis-HindIII and PA14_49890-F1-NheI and PA14_49890-R-cHis-HindIII , respectively ( Table S3 ) . PCR products were cloned into the NheI and HindIII sites of pMQ70 . The resulting plasmids , pMQ70-tpbA and pMQ70-tpbB , were transformed into PA14 and the mutants by conjugation . Briefly , 1 mL of overnight culture of the recipient strain ( PA14 or the mutant ) , helper strain ( HB101/pRK2013 ) , and donor strain ( TG1/pMQ70 , TG1/pMQ70-tpbA , or pMQ70-tpbB ) was washed with 1 mL of fresh LB medium . The mixture of three strains was incubated on LB plates at 37°C overnight . PA14 strains with pMQ70-based plasmid were selected on LB plates with 100 µg/mL rifampicin ( to kill the donor and helper ) , 300 µg/mL carbenicillin ( to kill P . aeruginosa without pMQ70-based plasmids ) , and 15 µg/mL gentamicin ( if a recipient was a PA14 mutant constructed using a transposon insertion with the GmR gene ) . If indicated , 0 . 05% arabinose was added to induce gene expression . Biofilm formation was examined in 96-well polystyrene plates using crystal violet staining [52] . Overnight cultures of P . aeruginosa were diluted to a turbidity of 0 . 05 at 600 nm with fresh LB medium , and then 150 µL of diluted bacterial culture was incubated in 96-well polystyrene plates for 2 , 4 , 8 , 24 , and 50 h . Ten wells were used for each strain and three independent cultures were used for each experiment . Trisodium orthovanadate , a tyrosine phosphatase-specific inhibitor , was added to LB medium at 10 mM . To observe colony morphology , overnight cultures were diluted to a turbidity of 0 . 005 at 600 nm with T-broth ( 10 g/L tryptone ) , and 2 µL of diluted cultures were spotted on Congo-red plates ( 10 g/L tryptone , 40 µg/mL Congo-red , and 20 µg/mL Coomassie brilliant blue ) [3] . Plates were incubated at 37°C or room temperature for 3 to 7 days . Swimming motility was examined with cells grown to a turbidity of 1 at 600 nm using 0 . 3% agar plates with 1% tryptone and 0 . 25% NaCl [53] and swarming motility was examined with BM-2 plates ( 62 mM potassium phosphate , 2 mM MgSO4 , 10 µM FeSO4 , 0 . 1% casamino acid , 0 . 4% glucose , and 0 . 5% Bacto agar ) [54] . Motility was measured after 24 h . Five plates were tested for each culture , and two independent cultures were used . The flgK [22] and rhlR [23] mutants were used as negative controls for swimming and swarming , respectively . Aggregation was examined by diluting overnight cultures with fresh LB medium in 5 mL screw-capped tubes from 0% ( no added fresh LB medium ) to 100% ( pure fresh LB medium ) . Cells were inverted gently several times and placed at room temperature for 15 min . Overnight cultures of PA14 , the tpbA mutant , and the pelA mutant were diluted to a turbidity of 0 . 005 at 600 nm in 4 mL T-broth , and the bacterial cultures were placed in a polycarbonate glass tube at 37°C or room temperature [3] . Pel-dependent EPS production was quantified as described previously [6] based on the amount of Congo red that binds to the EPS . Briefly , 1 mL of overnight culture was washed with 1 mL T-broth . Due to aggregative phenotype of the tpbA mutant , cell pellets of the tpbA mutant , wild-type , and pelA mutant ( negative control ) were sonicated three times at 3W for 10 sec . Bacterial suspensions in T-broth ( 500 µL ) were incubated with 40 µg/mL Congo-red at 37°C or room temperature with vigorous shaking . After 2 h , the absorbance of the supernatants of the each suspension was measured at 490 nm using a spectrophotometer . T-broth with 40 µg/mL Congo-red was used as a blank . Production of rhamnolipids was determined as described previously [55] . Overnight cultures were diluted to a turbidity of 0 . 05 at 600 nm in 25 mL LB medium and were re-grown at 250 rpm for 24 h to eliminate the effect of antibiotics . The supernatants of the bacterial cultures were used to determine the relative concentrations of rhamnolipids using orcinol/sulfuric acid . Rhamnose ( Fisher Scientific , Pittsburgh , PA ) was used as a standard . The P . aeruginosa genome array ( Affymetrix , P/N 510596 ) was used to investigate differential gene expression in biofilm cells between PA14 and the tpbA mutant . Biofilm cells were harvested from 10 g of glass wool [56] after incubation for 4 h and 7 h in LB with shaking at 250 rpm , and RNA was extracted with a RNeasy Mini Kit ( Qiagen ) [57]; note the RNase inhibitor RNAlater ( Applied Biosystems , Austin , TX ) was used for the 4 h and second 7 h set of microarrays . Global scaling was applied so the average signal intensity was 500 . The probe array images were inspected for any image artifact . Background values , noise values , and scaling factors of both arrays were examined and were comparable . The intensities of polyadenosine RNA controls were used to monitor the labeling process . If the gene with the larger transcription rate did not have a consistent transcription rate based on the 13 probe pairs ( p-value less than 0 . 05 ) , these genes were discarded . A gene was considered differentially expressed when the p-value for comparing two chips was lower than 0 . 05 ( to assure that the change in gene expression was statistically significant and that false positives arise less than 5% ) and when the expression ratio was higher than the standard deviation for the whole microarrays [58] , 1 . 4 for 4 h , 1 . 7 for the first 7 h replicate , and 2 . 2 for second 7 h replicate . All three sets of whole-transcriptome data were deposited at the Gene Expression Omnibus ( GSE13871 ) . qRT-PCR was performed using the StepOnePlus™ Real-Time PCR System ( Applied Biosystems , Foster City , CA ) . Expression of pelA , the PA4625 gene , and the PA4139 gene was determined using total RNA isolated from two independent biofilm cultures of PA14 and the tpbA mutant . The biofilm cells were grown and total RNA were isolated in the same manner as described above for the whole-transcriptome analysis . The primers for qRT-PCR are listed in Table S3 . The housekeeping gene rplU [44] was used to normalize the gene expression data . To isolate the suppressive loci for TpbA functions , a double mutant library was generated using the Tn5-luxAB transposon with the background of the tpbA mutation as described previously [59] . Briefly , 1 mL of overnight culture of the P . aeruginosa tpbA mutant and E . coli S17-1 ( λpir ) with Tn5-luxAB were grown on LB plates together overnight . Cells were harvested from the plate and resuspended in 10 mL of LB medium . Screening of cells with mutations in addition to tpbA was performed in three steps . Suppression of the highly-aggregative phenotype of the tpbA mutant was used first; the cell mixture ( P . aeruginosa single and double mutants along with E . coli S17-1 ) was placed at room temperature for 15 min and the supernatant was used for secondary screening ( cells with the tpbA mutant aggregative phenotype were therefore discarded ) . Supernatant cells were spread on Congo-red plates with 50 µg/mL gentamicin ( to kill E . coli ) , and 75 µg/mL tetracycline ( to kill the tpbA single mutant ) , and incubated for 3–4 days . P . aeruginosa double mutants with smooth surfaces were picked ( the tpbA mutant was red and wrinkled ) . The crystal violet biofilm assay was used for the third screening , and mutants showing decreased biofilm formation in comparison to that of the tpbA mutant were chosen as phenotype reversal mutants . The insertion position of Tn5-luxAB transposon was determined by two-step PCR as described previously [59] with primers LuxAB inside and Arb1 for the first round of PCR and LuxAB outside and Arb2 for the second round of PCR ( Table S3 ) . The PCR product was ligated into pGEM-T easy ( Promega , Madison , MI ) and sequenced using a BigDye Terminator Cycle Sequencing Kit ( Applied Biosystems , Foster City , CA ) . c-di-GMP was isolated as described previously [60] . P . aeruginosa was grown in 1 L of LB medium for 16 h at 250 rpm , and formaldehyde ( final concentration of 0 . 18% ) was added to inactivate degradation of c-di-GMP . Cells were harvested by centrifugation at 8 , 000 g for 10 min at 4°C . Nucleotide extract was prepared as described previously [60] . Cell pellets were washed with 40 mL of phosphate buffered saline ( pH 7 ) [61] with 0 . 18% formaldehyde and centrifuged at 8 , 000 g for 10 min at 4°C . The cell pellets were dissolved in H2O and boiled for 10 min . After cooling the samples on ice for 10 min , nucleotides were extracted in 65% ethanol . Supernatants were transferred , and the extraction was repeated . Pooled supernatants were lyophilized , and pellets were dissolved in 1 mL of 0 . 15 M triethyl ammonium acetate ( TEAA , pH 5 . 0 ) . The samples were filtered using a PVDF filter ( 0 . 22 µm ) , and 20 µL of each sample was fractionated using HPLC ( Waters 515 with photodiode array detector , Milford , MA ) with a reverse-phase column ( Nova-Pak® C18 column; Waters , 150×3 . 9 cm , 4 µm ) . Separations were conducted in 0 . 15 M TEAA at a 1 mL/min flow rate using gradient elution with acetonitrile ( 0% to 15% concentration ) . Synthetic c-di-GMP ( BIOLOG Life Science Institute , Bremen , Germany ) was used as a standard . The peak corresponding to c-di-GMP from the extract of the tpbA mutant was verified by co-elution with standard c-di-GMP . E . coli AG1/pCA24N-yddV that has an elevated c-di-GMP concentration [62] was also used as a positive control . To determine if TpbA is a phosphatase , tpbA was amplified with a Pfu DNA polymerase using primers PA14_13660-F-XbaI and PA14_13660-R-XhoI ( Table S3 ) . The PCR product was digested with XbaI and XhoI and was ligated in-frame to the polyhistidine tag sequence of the pET28b vector . The resulting plasmid , pET28b-13660c has the tpbA gene fused to a 6× His tag at the C-terminus ( TpbA-cHis ) and under control of the T7 promoter . The pET28b-13660c plasmid was confirmed by DNA sequencing with the T7 promoter and T7 terminator primers ( Table S3 ) . Production of TpbA-cHis was induced in E . coli BL21 ( DE3 ) cells with 1 mM IPTG at room temperature overnight . TpbA-cHis was purified using a Ni-NTA resin ( Qiagen , Valencia , CA ) as described in a manufacturer's protocol . Purified TpbA-cHis was dialyzed against buffer ( 50 mM Tris-HCl , 100 mM NaCl , 10% glycerol , 0 . 01% Triton X-100 , pH 7 . 5 ) at 4°C overnight . The p-nitrophenyl phosphate assay ( pNPP ) was used to examine TpbA-cHis phosphatase activity [63] . Purified TpbA-cHis protein was incubated in 100 µL of reaction buffer ( 50 mM Tris-acetate , 10 mM MgCl2 , 10 mM pNPP , 5 mM DTT , pH 5 . 5 ) at 37°C for 1 h . The reaction was quenched by adding 900 µL of 1 M NaOH . Trisodium orthovanadate , a specific inhibitor for tyrosine phosphatase [30] , was used at 10 mM . p-nitrophenol was measured at an absorbance of 405 nm . An extinction coefficient of 1 . 78×104 M−1 cm−1 was used to calculate the concentration of p-nitrophenol . To examine if TpbA is a tyrosine specific phosphatase , a tyrosine phosphatase assay was performed using the Tyrosine Phosphatase Assay System ( Qiagen ) . Eight micrograms of TpbA-cHis were incubated with either 50 µM phosphotyrosine peptide type I ( END ( pY ) INASL ) or peptide type II ( DADE ( pY ) LIPQQG ) in a reaction buffer ( 50 mM Tris-Acetate , 10 mM MgCl2 , pH 5 . 5 ) at 37°C for 3 h . The reaction was quenched using a molybdate dye solution and incubated for 30 min at room temperature . Released phosphate was quantified by measuring the absorbance at 630 nm . TpbA-cHis protein was expressed in BL21 ( DE3 ) cells with 1 mM IPTG for 4 h at 37°C . Periplasmic proteins were purified using a PeriPreps Periplasting Kit ( Epicentre Technologies , Madison , WI ) as well as cytoplasmic and membrane proteins . Escherichia coli CpdB [64] was used as a control of periplasmic protein and the E . coli OxyR [65] was used for the cytoplasmic control . Fractionated proteins as well as TpbB were analyzed by 12% SDS-PAGE . Site-directed mutagenesis of the predicted periplasmic tyrosine residues of TpbB was performed to convert them to phenylalanine ( Y48F , Y62F , and Y95F ) ; it was reasoned that phenylalanine would provide a similar bulky side chain but remove the hydroxyl moiety needed for phosphorylation [66] . The mutations were introduced into pMQ70-tpbB using Pfu DNA polymerase and QuikChange Site-Directed Mutagenesis Kit ( Stratagene , La Jolla , CA ) , and the primers are listed in Table S3 . The resulting plasmids , pMQ70-tpbB-Y48F , pMQ70-tpbB-Y62F , and pMQ70-tpbB-Y95F were transformed into the tpbB mutant by conjugation and aggregation was assayed . DNA sequencing was used to confirm the tyrosine mutations and that no other mutations were introduced into the promoter or protein-coding sequences . The promoter region of tpbA ( ptpbA ) , including 399 bp upstream of the start codon and 31 bp of the open reading frame , was amplified using Pfu DNA polymerase with primers pPA14_13660-F-HindIII and pPA14_13660-R-BamHI ( Table S3 ) . The PCR product ( 430 bp ) was cloned into the HindIII/BamHI sites of pLP170 to produce pLP-ptpbA , and it was conjugated into PA14 and the QS-related mutants using helper strain HB101/pRK2013 [67] , [68] . Transformants were grown overnight in LB medium with 300 µg/mL carbenicillin , reinoculated at a turbidity of 0 . 05 at 600 nm , and grown for another 6 h . Biofilm cells were harvested from 4 g of glass wool after incubation for 6 h in LB at 37°C with shaking at 250 rpm . β-galactosidase activity was measured using suspension cells and biofilm cells as described previously [69] . Similarly , β-galactosidase activity of plasR::lacZ ( pPCS1001 ) and prhlR::lacZ ( pPCS1002 ) was examined in PA14 and the tpbA mutant . | Most bacteria live in biofilms , which are complex communities of microorganisms attached to a surface via polysaccharides; these biofilms are responsible for most human bacterial diseases . The pathogen Pseudomonas aeruginosa is best-studied for biofilm formation . Currently , it is recognized that cell communication or quorum sensing is important for biofilm formation , but how these external signals are converted into internal signals to regulate the networks of genes that result in biofilm formation is not well understood . Here , by studying 5850 bacterial strains , each of which lacks a single protein , we identify a new enzyme of P . aeruginosa , a tyrosine phosphatase ( TpbA ) , that links extracellular quorum sensing signals to polysaccharide production and biofilm formation . We find that TpbA is subject to control by quorum sensing signals , that it is in the periplasm , and that it controls the level of the intracellular secondary messenger 3 , 5-cyclic diguanylic acid ( c-di-GMP ) . By controlling c-di-GMP concentrations , TpbA serves to regulate biofilm formation , rapid cell movement on the surface , colony morphology , cell aggregation , and polysaccharide production . The importance of our study is that it shows the secondary messenger c-di-GMP may be regulated by tyrosine phosphorylation; hence , it provides a new target for controlling bacterial social behavior . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"molecular",
"biology/post-translational",
"regulation",
"of",
"gene",
"expression",
"genetics",
"and",
"genomics/gene",
"discovery",
"cell",
"biology/cell",
"signaling",
"genetics",
"and",
"genomics/gene",
"expression",
"cell",
"biology/microbial",
"growth",
"and",
"development",
"genetics",
"and",
"genomics/gene",
"function",
"cell",
"biology/extra-cellular",
"matrix",
"infectious",
"diseases/bacterial",
"infections",
"microbiology/microbial",
"growth",
"and",
"development",
"cell",
"biology/cell",
"adhesion",
"cell",
"biology/gene",
"expression"
] | 2009 | Connecting Quorum Sensing, c-di-GMP, Pel Polysaccharide, and Biofilm Formation in Pseudomonas aeruginosa through Tyrosine Phosphatase TpbA (PA3885) |
Many complex networks such as computer and social networks exhibit modular structures , where links between nodes are much denser within modules than between modules . It is widely believed that cellular networks are also modular , reflecting the relative independence and coherence of different functional units in a cell . While many authors have claimed that observations from the yeast protein–protein interaction ( PPI ) network support the above hypothesis , the observed structural modularity may be an artifact because the current PPI data include interactions inferred from protein complexes through approaches that create modules ( e . g . , assigning pairwise interactions among all proteins in a complex ) . Here we analyze the yeast PPI network including protein complexes ( PIC network ) and excluding complexes ( PEC network ) . We find that both PIC and PEC networks show a significantly greater structural modularity than that of randomly rewired networks . Nonetheless , there is little evidence that the structural modules correspond to functional units , particularly in the PEC network . More disturbingly , there is no evolutionary conservation among yeast , fly , and nematode modules at either the whole-module or protein-pair level . Neither is there a correlation between the evolutionary or phylogenetic conservation of a protein and the extent of its participation in various modules . Using computer simulation , we demonstrate that a higher-than-expected modularity can arise during network growth through a simple model of gene duplication , without natural selection for modularity . Taken together , our results suggest the intriguing possibility that the structural modules in the PPI network originated as an evolutionary byproduct without biological significance .
Many complex networks are naturally divided into communities or modules , where links within modules are much denser than those across modules [1] ( Figure 1 ) . For example , human individuals belonging to the same ethnic groups interact more than those from different ethnic groups [2] . Studying the modularity of a network not only provides structural information about the network , but may also reveal the underlying mechanisms that determine the network structure . The concept of modularity is not new to biologists . In fact , cellular functions are widely believed to be organized in a highly modular manner , where each module is a discrete object composed of a group of tightly linked components and performs a relatively independent task [3–7] . It is interesting to examine whether this modularity in cellular function arises from modularity in molecular interaction networks such as the transcriptional regulatory network and protein–protein interaction ( PPI ) network . Many authors have attempted to separate modules in the PPI network based on either the network topology alone or with additional information about gene function and expression [8–16] . They generally report high modularity in the PPI network , with evidence for a rough correspondence between PPI modules and functional units . All these analyses , however , suffered from a serious bias in the current PPI data . The PPI data include binary interaction information that is either directly obtained from experiments such as the yeast two-hybrid ( Y2H ) assay [17 , 18] , or indirectly inferred from stable protein complexes [19] . High-throughput protein complex identification is usually mass-spectrometry–based [20–23] ( e . g . , tandem-affinity purification ) . These methods involve the discovery of a complex of interacting proteins including a tagged bait protein , but do not provide information about direct pairwise protein–protein interactions [19 , 24] . Some small-scale biochemical methods , such as co-immunoprecipitation [25] and affinity precipitation [26] , can also identify protein complexes without providing pairwise protein interaction information . Protein complex data obtained by one of these methods are then translated into binary PPIs by either the “matrix” or the “spoke” model [19] ( Figure 2 ) . The matrix model assumes that all members of a protein complex interact with each other , whereas the spoke model assumes that all nonbait members of a complex interact with the bait . It is obvious that use of the matrix model creates PPI modules corresponding to protein complexes . The spoke model can also affect modularity because the bait is interpreted by the model as a hub ( i . e . , a highly connected node ) , while in reality it may not be a hub . Because the reliability of the two models is unknown , it is possible that the prevailing modularity of PPI networks is an artifact of these models . In this work , we explore the above possibility by analyzing the modularity of two yeast PPI networks . The first is referred to as the PIC network , as it is the PPI network including protein complex data , whereas the second is named the PEC network , as it the PPI network excluding all edges inferred from protein complexes . Because we are assessing the modularity of the PPI network per se , only the network topology will be used in separating modules . Our analyses show that although both PIC and PEC networks are highly modular , the identified modules lack obvious correspondence to functional units and are not evolutionarily conserved . We use computer simulation to show that modularity can arise in a simple model of network growth through gene duplication , without the involvement of selection for modularity . Together , our findings suggest that structural modules in PPI networks may have arisen as an evolutionary byproduct without biological significance .
We downloaded the PPI data for the budding yeast Saccharomyces cerevisiae from the Munich Information Center for Protein Sequences ( MIPS ) [27] . The dataset was human-curated and contained mostly binary interactions directly observed in Y2H experiments . In addition , about 10% of the binary interactions in the dataset were inferred using either the spoke or matrix model from protein complexes identified by high-confidence small-scale experiments . This entire dataset is referred to as the PIC network here . Based on the MIPS annotation , we removed from the PIC network those binary interactions that were inferred from protein complexes , resulting in the PEC network . Because it is only meaningful to separate modules within a connected part of a network , we studied the largest connected subset ( i . e . , the giant component ) , of a network . The giant component contains more than 90% of all nodes in the yeast PPI network . For simplicity , we refer to the giant component of a network as the network , unless otherwise noted . Table 1 lists some important parameters for the PIC and PEC networks studied here . The extent of modularity for a particular modular separation of a network is often measured by M = [ ( ls/L ) − ( ks/2L ) 2] , where N is the number of modules , L is the total number of edges in the network , ls is the number of edges within module s , and ks is the sum of the degrees of the nodes in module s [28 , 29] . The degree of a node is simply the number of edges that the node has . The particular separation that maximizes M is considered the optimal modular separation and the corresponding M is referred to as the modularity of the network ( Figure 1 ) . In essence , M is the difference between the observed and expected proportions of within-module edges in the network . Here , the expected proportion is computed from a nonmodular network where edges are equally likely to be within and between modules . Several algorithms are available to separate a network into modules and obtain the maximal M . Empirical and simulation studies showed that the method of Guimera and Amaral [28] has the best performance because it can give the most accurate module separation and highest M [30] . We therefore used this method to separate modules in the yeast PIC and PEC networks . To obtain the highest M , we used delicate parameter settings in the simulated annealing algorithm . It took a typical desktop computer ~3 d to separate a yeast PPI network . The PIC network is separated into 26 modules with a modularity of 0 . 6672 , while the PEC network is divided into 22 modules with a modularity of 0 . 6583 ( Table 1 ) . The density ratio , defined by the ratio of the number of within-module edges to the number of between-module edges , is only slightly lower for PEC than for PIC networks ( Table 1 ) . A random network may also have a nonzero modularity by chance or due to certain degree distributions [31] . Also , the modularity values of two networks with different sizes or different average degrees cannot be compared directly [31] . Thus , to measure the modularity of a network , we compare it with a random network of the same size and same degree distribution , which is generated by the local rewiring algorithm [32] . To speed up the computation , we used moderate parameter settings and faster runs ( ~4 h per network ) to estimate modularity . For the yeast PIC network , the modularity for 500 randomly rewired networks has a mean of 0 . 5466 and a standard deviation of 0 . 0023 , while the real PIC network has a modularity of 0 . 6555 under this parameter setting ( Figure 3A ) . We use z-score , or the number of standard deviations higher than the random expectation to measure the deviation of the modularity of a network from its random expectation . This z-score , referred to as the scaled modularity to differentiate it from z-scores of other properties , is ( 0 . 6555 – 0 . 5466 ) /0 . 0023 = 47 for the PIC network . Under the same parameter setting , the modularity for the real PEC network is 0 . 6481 . The modularity for 500 randomly rewired PEC networks has a mean of 0 . 5764 and a standard deviation of 0 . 0027 ( Figure 3B ) . In other words , the scaled modularity for the PEC network is ( 0 . 6481 − 0 . 5764 ) /0 . 0027 = 27 . Thus , both PIC and PEC networks show significantly greater modularity than randomly rewired networks . As expected , the scaled modularity of PIC is much greater than that of PEC . This difference is largely due to the exclusion of protein complex data in the PEC network . In fact , when we randomly removed 10% of edges from the PIC network , the scaled modularity decreased only slightly ( from 47 to 42 ) . Given the substantive difference in scaled modularity , PIC and PEC networks should also differ in the compositions of their modules . We measured the similarity in module composition between different separations of the same network ( or shared nodes in the case of different networks ) by the normalized mutual information ( NMI ) index [30] . A higher NMI indicates a higher similarity in module composition . The NMI between the PIC network and PEC network is 0 . 35 . As a control , we measured the NMI between the PIC network and a reduced network generated by random removal of 10% of the edges in PIC . This control NMI has a mean of 0 . 41 and a standard deviation of 0 . 018 ( from 200 replications ) . Thus , the NMI between PIC and PEC is significantly lower than that between PIC and its randomly reduced networks ( p < 0 . 002 ) ( Figure 3C ) . Because simulated annealing is a stochastic algorithm , different runs may yield slightly different partitions . We thus separated modules in PIC and PEC networks with different random seeds 50 times , and these replications confirmed that the above finding of a lower NMI between PIC and PEC than by chance is genuine ( p < 10−10 , Mann-Whitney U test ) . Together , these analyses demonstrate that the inclusion of interactions inferred from protein complexes in the PPI network has a great impact on network modularity . Because we identified the PPI modules based entirely on the topology of the network , it is important to ask whether such structural modules correspond to functional units . To address this question , we utilized the functional annotation of yeast genes in the CYGD database [33] . At the highest level of annotation , each yeast gene is classified into one or several of 17 functional categories ( Figure 4 ) . If the structural modules correspond to functional units , we should expect a nonrandom among-module distribution of the genes of a given functional category . For example , in the PIC network , there are 361 genes belonging to functional category A ( cell type differentiation; see Figure 4A ) . A χ2 test showed that these genes are not randomly distributed across the 26 PIC modules ( χ2 = 317 , df = 25 , p < 10−5; see the circles in Figure 4A ) . This test was conducted for each functional category , and almost all functional categories showed significant nonrandom distributions across PIC modules ( even after considering multiple testing ) . In contrast , the PEC network has fewer functional categories showing significant nonrandom distributions . This trend is particularly evident at the highest level of statistical significance ( six categories in PEC versus 14 in PIC ) ( Figure 4B ) . If structural modules correspond to functional units , we also expect that the majority of genes in a module belong to only one or a few functional categories . In other words , each module should have one or a small number of overrepresented functional categories . Testing this prediction is not easy because one gene may belong to multiple functional categories . We thus used computer simulations . For example , module 1 of the PIC network comprises 227 proteins , 92 of which belong to functional category A ( Figure 4A ) . We randomly chose 227 genes from the network and counted the number of category A genes . We repeated this procedure 100 , 000 times to estimate the probability that the number of category A genes in the randomly picked 227 genes is equal to or greater than 92 . This probability is indicated with different colors in the small squares of Figure 4A . Because 17 functional categories were tested for each module , to control for multiple testing we used 10−3 as the cutoff for statistical significance for each category . It can be seen that in 16 ( 62% ) of the 26 PIC modules , at least one functional category is enriched . In comparison , only 7 ( 32% ) of the 22 PEC modules have at least one enriched functional category . The above difference between PIC and PEC modules is statistically significant ( p < 0 . 05 , χ2 test ) . The two analyses above revealed nonrandom distributions of protein functions across structural modules . To quantitatively measure how well structural modules correspond to functional units , we used a correlation analysis . For a pair of proteins from a PPI network , we ask if they belong to the same module ( co-membership ) and if they belong to the same functional category ( co-functionality ) . Two proteins are considered to possess co-functionality as long as they share at least one function . If structural modules correspond well to functional units , protein pairs within the same module should share function whereas protein pairs across modules should not share function . In other words , we should observe a strong positive correlation between co-membership and co-functionality of protein pairs . We enumerated all possible protein pairs and found the correlation to be statistically significant in both PIC ( p < 10−300 ) and PEC ( p < 10−100 ) networks . However , the level of correlation is extremely low in both PIC ( r2 = 0 . 0813% ) and PEC ( r2 = 0 . 00675% ) networks ( Figure 4C and 4D ) , indicating that less than 0 . 1% of the variance in protein-pair co-membership is explainable by co-functionality . We also found that the r value for PEC is significantly lower than that for PIC when we repeated module separations 50 times with different random seeds ( p < 10−5 , Mann-Whitney U test ) . The observation of a low level of correlation is not due to the presence of many multifunctional proteins , because the low correlation is also observed even when we consider only monofunctional proteins ( r2 = 0 . 0384% and p < 10−37 for PIC; r2 = 0 . 0331% and p < 10−30 for PEC ) . Hence , although there is significant non-randomness in protein functions across structural modules , the correspondence between structural modules and functional units is extremely weak in both PIC and PEC networks , especially in the latter . We also examined the cellular locations of each protein [34] and tested whether members of a structural module tend to be co-localized , as would be expected if structural modules represent functional units . Our results were generally similar to those for functional categories . Although some nonrandom patterns were observed , the correspondence between structural modules and cellular locations is extremely weak in both PIC and PEC networks , especially in the latter ( Figure S1 ) . If a structurally defined PPI module represents a functional unit , the composition of the module should be evolutionarily conserved . To test this prediction , we applied the same module separation algorithm to the fruit fly ( Drosophila melanogaster ) PPI network , which was constructed from binary PPIs obtained in high-throughput Y2H experiments [35] . Because the fly data do not contain any interactions inferred from protein complexes , we expect that the fly PPI network behaves more similarly to the yeast PEC network than to the PIC network . We thus examine the evolutionary conservation of modular structures between the yeast PEC network and the fly network . We separated the fly network into 27 modules , with a modularity of 0 . 6851 and a scaled modularity of 29 ( Table 1 ) . Hence , the scaled modularity of the fly network is comparable to that of the yeast PEC network ( 27 ) . There are 691 orthologous proteins between the giant component of the yeast PEC network and that of the fly network . We here again use NMI to measure the similarity in module compositions between two networks . The NMI value between the yeast PEC and fly PPI networks is 0 . 14 . If the modular structures are evolutionary conserved between the two networks , the above NMI value should be significantly greater than that between the actual fly network and a randomly separated yeast network . We randomly separated the yeast PEC network into 26 modules by conserving the actual module sizes and then computed NMI between the real fly modules and the randomly separated yeast modules . To make this comparison , we repeated this process 10 , 000 times and obtained the frequency distribution of NMI ( Figure 5A ) . The observed NMI between the real fly and real yeast networks falls in the central part of the distribution , indicating that the yeast and fly modules are no more similar to each other than by chance ( p > 0 . 6 ) and revealing a complete lack of evolutionary conservation in PPI modules between the two species . Because modular structures are often hierarchically organized [7] , it is possible that a low level of structure is evolutionarily conserved despite the lack of conservation at the whole-module level . Pairwise relationships between proteins represent the lowest possible structure in the PPI network . We invented a conservation index for pairs of proteins ( CIP ) . Between species X and Y , CIP is defined as the probability that the Y orthologs of two X proteins belonging to the same module in X also belong to the same module in Y . CIP is 0 . 048 between the yeast and fly , which is not significantly different from the expectation derived by comparison of the fly network to a random separation of the yeast network ( p > 0 . 6; 10 , 000 simulations; Figure 5B ) . Thus , even at the lowest structural level , yeast and fly modules are not evolutionarily conserved . Note that CIP measures the conservation of co-membership in a module between two proteins , regardless of whether these two proteins interact with each other . CIP does not measure the conservation of PPIs . If two yeast proteins engage in a PPI and their respective fly orthologs also engage in a PPI , these two PPIs are referred to as orthologous PPIs [36] . Between the yeast PEC and fly PPI networks , there are 45 orthologous PPIs . In comparison , between the fly network and 1 , 000 randomly rewired yeast networks ( with the degree of each node unchanged ) , there are only 0 . 58 orthologous PPIs on average ( standard deviation = 0 . 75 ) . Thus , orthologous PPIs are evolutionarily conserved between the two species . We also examined the evolutionary conservation of structural modules between yeast and the nematode Caenorhabditis elegans . Although the PPI data for C . elegans are highly incomplete , with only 2 , 387 proteins and 3 , 825 interactions in the giant component , the results we obtained ( Figure S2 ) are similar to those from the comparison between yeast and fruit fly networks . If structural modules represent functional units , proteins with links to many modules should be evolutionarily more conserved than those with links largely within a module , because multifunctional or pleiotropic proteins tend to be conserved [37 , 38] . Guimera and Amaral [28] defined the participation coefficient of a node by PC = 1 − ( ki/k ) 2 , where k is the degree of the node , ki is the number of links from the node to any nodes in module I , and N is the total number of modules . A high PC indicates that a node participates in the functioning of many modules . These authors found that the propensity for an enzyme gene to be lost during evolution is negatively correlated with the PC of the enzyme in the metabolic network [28] . Such an observation strongly suggests that the modular structure in the metabolic network has biological significance . It is therefore useful to examine PC for the proteins in the PPI network . It has previously been debated whether the degree of a protein in the PPI network influences its evolutionary rate [39–44] . Because past studies did not exclude PPIs inferred from protein complexes , it is possible that some of previous results were due to artifacts of such inferences . Separate analyses of the PIC and PEC networks may help to answer this question . We first measured the rate of protein evolution by the number of nonsynonymous nucleotide substitutions per nonsynonymous site ( dN ) between orthologous genes of yeast species S . cerevisiae and S . bayanus . We chose this species pair because their divergence level is appropriate for obtaining informative and reliable dN estimates [45] . We found that the dN of a protein is significantly negatively correlated with its total degree in the yeast PIC network ( p < 0 . 001; Table 2 ) , but not with its degree in the PEC network ( p > 0 . 4 ) . Thus , when protein complexes are not considered , there is no significant correlation between dN and degree . When we separated the links of a node into within-module links and between-module links , we found a significant correlation between dN and the within-module degree ( i . e . , the number of within-module links ) in the PIC network . This correlation is again absent in the PEC network , suggesting that the correlation between dN and within-module degree is largely attributable to protein complexes . In neither the PIC nor the PEC network did we find a significant correlation between dN and the between-module degree ( i . e . , the number of links across modules ) . Similar results were found between dN and PC of a protein ( Table 2 ) . Furthermore , even when we divided the proteins into different topological roles by their PCs and degrees , as was done by Guimera and Amaral for the metabolic network , no significant correlation between these roles and dN was observed ( Table 2 , bottom two rows ) . We also measured the rate of protein evolution by the propensity for gene loss ( PL ) across 12 fungal species whose draft genome sequences are available . The results obtained for PL are qualitatively similar to those for dN ( Table 2 ) . Taken together , there is no observable impact of the within-module , between-module , or total PPI degree of a protein on its evolutionary rate when protein complexes are excluded . Furthermore , if structural modules correspond to functional units , a protein with higher participation in various modules should be more pleiotropic ( or multifunctional ) and thus should be more conserved in evolution [37 , 38] . However , we found no impact of the extent of participation in various modules on the evolutionary rate of a protein . This negative result is consistent with the idea that structural modules do not correspond to functional units . The growth rate of a yeast strain with a gene deleted can measure the importance of the gene under the tested condition . Growth rate is known to be negatively correlated with the PPI degree of a gene [39 , 46–48] . We confirmed this result in both PIC and PEC networks , although the significance is only marginal in the latter ( Table 2 ) . Interestingly , for both networks , this significance is also found for within-module degrees , but not for between-module degrees . This phenomenon may arise because the between-module degree is often much smaller ( mean = 1 . 04 for PIC and 0 . 98 for PEC ) than the within-module degree ( mean = 2 . 70 for PIC and 2 . 48 for PEC ) and thus contributes less to the total degree of a node . Growth rate also contains the information of gene essentiality , as essential genes have zero growth rates whereas nonessential genes have positive growth rates . Thus , similar results are obtained when we analyze the genes by gene essentiality rather than by growth rate . Because both PIC and PEC networks have significantly higher modularity than that of their randomly rewired networks but the identified modules exhibit little biological significance , it is puzzling how the modular structure could have arisen in evolution . Earlier studies suggested that modularity can originate by gene duplication [49 , 50] . However , in these studies modularity is defined by hierarchical clustering or a clustering coefficient , which lacks an objective function to identify the best module separation and to compute network modularity . We thus conducted computer simulations to examine whether the network modularity as defined in this paper can arise from evolution by gene duplication . Because duplication–divergence models can generate many network features similar to real PPI networks [50 , 51] and have clear biological bases [52–54] , we simulated network growth by a duplication–divergence model starting from a pair of connected nodes . Briefly , at each step , a node ( A ) is randomly picked and duplicated along with all its edges to generate its paralogous node ( A′ ) . We refer to two edges , one from A and the other from A′ , as a pair of edges if they both end at the same third node . To simulate functional divergence after gene duplication , we randomly remove one edge from each pair of edges , until A and A′ share 90% of edges . This duplication–divergence process was repeated 300 times to generate a network of 302 nodes . The resulting network has 212 nodes in its giant component ( Table S1 , first row ) . We found the modularity and scaled modularity of this simulated network to be 0 . 6717 and 29 , respectively ( Figure 6; Table S1 ) . We conducted ten simulation replications , and all cases show similarly high modularity and scaled modularity that are comparable with those of the yeast and fly PPI networks ( Table S1 ) . In fact , we found that many different combinations of simulation parameters can give rise to modular networks , and the specific model of evolution by gene duplication ( e . g . , the subneofunctionalization model [52] ) does not appear to matter much to the result of high modularity ( unpublished data ) . Although self-interactions can be biologically important , they are not considered in our simulation because such interactions are disregarded in the module separation algorithm of Guimera and Amaral [28] .
In this work , we conducted a comprehensive analysis of modular structures in yeast protein interaction networks . Rather than lumping binary PPIs directly observed in experiments with those indirectly inferred from protein complexes , we separately analyzed the PIC network , which includes inferred binary PPIs , and the PEC network , which excludes inferred binary PPIs . This distinction is necessary because inferences of binary interactions from protein complexes introduce errors to the network structure , which hamper accurate measurement of network modularity . Given that protein complexes likely represent true ( functional ) modules in the network , the unanswered question is whether the network structure is still modular when the PPIs inferred from protein complexes ( ~10% in our PIC network ) are removed . We found that both PIC and PEC networks are significantly more modular than expected by chance , the scaled modularity of the PIC network is substantively greater than that of the PEC network , and the module compositions of the two networks are significantly different . The latter two results are expected , because the current models for inferring binary PPIs from protein complexes tend to increase modularity . Consistent with these results , we found that the fruit fly PPI network , which is entirely based on experimentally determined binary PPIs , has a comparable scaled modularity to that of the yeast PEC network . In spite of the presence of significant modularity in the yeast PEC network , the identified structural modules do not appear to correspond to functional units . This is reflected in three analyses . First , for some functional categories , their member genes are distributed randomly among structural modules . Second , for most structural modules , there are no enriched functional categories . Third , for protein pairs , the correlation ( r2 ) between co-membership in a module and co-functionality , although significantly greater than zero , is lower than 0 . 1% . Our results contradict several previous studies which claimed that PPI modules correspond well to functional units [8–16] . This difference is in part owing to the inclusion of protein complexes in these early studies . Furthermore , some studies utilized more than the PPI network topology in separating modules . For example , Tornow and Mewes considered gene co-expression patterns [15] . Although such practices may help identify functional modules , they do not objectively evaluate whether the PPI network itself has a biologically meaningful modular structure . Many studies also suffered from the lack of an efficient algorithm to identify the maximum modularity , resulting in suboptimal modular separations with many small modules . For example , Pereira-Leal and colleagues [13] separated the yeast PPI network into 1 , 046 modules , with an average size of eight proteins per module . A small module may appear to have a better functional correspondence than a large module , because the chance probability of functional similarity among a few proteins is considerably greater than that among a large number of proteins . Because the module separation algorithm we used here is superior to the earlier algorithms [30] , under the same definition of modularity our results are expected to be more reliable than those based on inferior algorithms . Although many authors have claimed that PPI networks are modular with significant functional correspondence , none have examined the evolutionary conservation of PPI modules . By comparing the yeast PEC network and fly PPI network , we found that PPI modules are not more conserved than the chance expectation at the whole-module level . Furthermore , even at the protein–pair level , the PPI modules are not more conserved than by chance . These findings are consistent with our observation of minimal correspondence between yeast PEC modules and functional units . Interestingly , PPIs are found to be conserved between the yeast and fly , suggesting that the lack of conservation of modules cannot be trivially explained by the lack of conservation of individual interactions in the network . The participation coefficient of a node measures the extent of the distribution of links from the node to all modules . If PPI modules correspond to functional units , proteins with high participation coefficients should have higher degrees of pleiotropy ( or multifunctionality ) and be more conserved than those with low participation coefficients , because pleiotropic or multifunctional proteins are known to be evolutionarily conserved [37 , 38] . This correlation was not observed in either the PIC or PEC network when either dN or PL was used as a measure of a protein's evolutionary rate . Thus , the results again point to the lack of correspondence between PPI modules and functional units . Taken together , our analyses strongly suggest that the yeast PEC network has a modular structure , which , nevertheless , lacks detectable biological significance . One may argue that the PEC network actually contains biologically important structural modules , but such modules are difficult to identify due to the incompleteness and inaccuracy of current PPI data . While this possibility cannot be entirely ruled out , we note that the PPI data we used here are generally regarded as of relatively high quality [27] . Furthermore , according to recent estimates , our PPI data should cover 25% to 50% of all PPIs in the yeast interactome [24 , 55] . Several observations , such as the negative correlation between the growth rate of a single-gene deletion yeast strain and the PPI degree of the gene ( Table 2 ) , suggest that the current PPI data contain biologically meaningful signals . An alternative explanation of the PPI modularity that lacks biological significance is that modularity may be an evolutionary byproduct . Inspired by earlier studies [49 , 50] , we demonstrate by computer simulation that a simple model of gene duplication–divergence can generate networks with a scaled modularity comparable to that observed in the yeast and fly PPI networks . This result suggests that the modularity in the PPI networks may indeed have no biological significance and has not been under selection . Because gene duplication is the primary source of new genes and new gene functions [56] , our simulation is biologically relevant . It is possible that evolutionary processes other than gene duplication also contributed to the origin of network modularity . For example , if assortative links ( i . e . , links between nodes of similar degrees ) are disfavored , as has been observed in PPI networks [57 , 58] , modularity may arise . PPI networks also have clustering coefficients higher than chance expectation , meaning that two proteins that both interact with the third one also tend to interact with each other [3] . Natural selection for higher clustering coefficients for some nodes of the network may also raise modularity . It has been intensely debated to whether there is a negative correlation between the PPI degree of a protein and the evolutionary rate ( dN ) of the protein [39–44] . We found this correlation to be statistically significant for the PIC network , but not significant for the PEC network . These observations suggest that the significant correlation is simply due to lower evolutionary rates for proteins involved in protein complexes than those not involved in complexes . Our result is consistent with a recent study reporting the lack of a significant correlation when PPIs were curated from literature [39] . Because proteins involved in complexes tend to have exceptionally high degrees as a result of indirect inference of PPIs by the matrix or spoke model , our result is also consistent with the finding that only the most prolific interactors tend to evolve slowly [44] . Recently , Han and colleagues [59] classified hubs ( i . e . , high-degree nodes ) in the PPI network into party hubs and date hubs . The former are those proteins whose interaction partners have similar expression profiles across various conditions , whereas the latter are those whose partners have different expression profiles . Party hubs have been interpreted as proteins that function within a biological process ( or a functional module ) , whereas date hubs are thought to link different functional modules . Fraser reported that party hubs are evolutionarily more conserved than date hubs , and suggested that this pattern may reflect a tendency for evolutionary innovations to occur by altering the proteins and interactions between rather than within modules [60] . A closer examination of the party hubs and their partners reveals that the majority of them form protein complexes , whereas date hubs and their partners do not form complexes . Thus , Fraser's observation is explainable by a lower evolutionary rate of proteins involved in complexes than those not in complexes , without invoking additional evolutionary forces . PPI networks have been subject to many structural , functional , and evolutionary analyses in the past few years . Our results show that removing a small fraction ( ~10% ) of PPIs that are inferred from protein complexes can have a substantial effect on the analysis . This observation raises a warning about many results regarding PPI networks , because they have usually been based on the PIC network that contains many potentially false PPIs inferred for members of protein complexes . As such false interactions are not randomly distributed in the network , their potential detrimental effect is particularly alarming . The PIC data we used do not contain high-throughput protein complex data such as those in [21 , 22] . In many PPI databases , such as BIND [61] , DIP [62] , and the new literature-curated dataset [63] , about half or more of the PPIs are inferred from protein complexes . The recent genome-wide surveys of all protein complexes in the yeast added even more complexes to the PPI data . Inclusion of inferred PPIs from these complexes would affect the network structure even more . We caution that use of such PPI data may produce misleading results . Systems biology is a nascent field with many hopes as well as much hype [64] . It has been of great interest to identify nonrandom topological structures such as motifs and modules in molecular networks [5 , 28 , 65] . Such nonrandom patterns are often interpreted as having functional significance and having been particularly favored by natural selection [28 , 66 , 67] . While this may be true in many cases , a nonrandom network structure can also originate as a byproduct of other processes without having its own function . Recent studies suggested that motifs in transcriptional regulatory networks do not represent functional units and are not subject to natural selection [68] . Rather , random gene duplication and mutation could give rise to motifs [69] . A recent study even suggested that the high abundance of feed forward loops in regulatory networks could be an evolutionary byproduct [70] . Our results add yet another network structure that is widely believed to be of great biological importance to this growing list of potential evolutionary byproducts . That being said , the modular organization of cellular functions is real , and whether this organization is also an evolutionary byproduct or has been actively selected for remains to be scrutinized .
The budding yeast ( S . cerevisiae ) PPI data were from the MPact dataset [27] of MIPS ( ftp://ftpmips . gsf . de/yeast/PPI/PPI_18052006 . tab ) , which contains human-curated high-throughput and small-scale binary interactions directly observed in experiments , as well as binary interactions inferred from high-confidence protein complex data . Only nonself physical interactions were considered . After excluding PPIs involving mitochondrial genes , we built the PPI network named PIC ( PPI including protein complexes ) . The giant component of the PIC network is composed of 3 , 886 proteins linked by 7 , 260 nonredundant interactions . To build the PEC ( PPI excluding protein complexes ) network , we retained only those binary interactions in the PIC network that had direct experimental evidence . The giant component of the PEC network contains 3 , 696 proteins linked by 6 , 403 interactions . The fruit fly ( Drosophila melanogaster ) PPI data came from [35] ( http://www . bme . jhu . edu/labs/bader/publications/giot_science_2003/flyconf . txt ) . A moderate confidence level ( 0 . 25 ) was chosen to generate the fly PPI network with a comparable average degree to the yeast PEC network . In total , the giant component of the fly PPI network contains 6 , 280 proteins linked by 10 , 210 interactions , all generated by Y2H experiments . The nematode ( C . elegans ) PPI data were from [71] ( http://vidal . dfci . harvard . edu/interactomedb/WI5 . txt ) . Only the PPIs identified by Y2H experiments are used . In total , the nematode PPI network contains 2 , 624 proteins and 3 , 967 interactions , of which 2 , 387 proteins and 3 , 825 interactions are in the giant component . We used the yeast functional annotations in the CYGD database [33] ( ftp://ftpmips . gsf . de/yeast/catalogues/funcat/funcat-2 . 0_data_18052006 ) , considering only the highest level of annotation . Functional categories containing <15 proteins and the category of unknown functions were removed . The cellular localization data for yeast proteins were from [34] ( http://yeastgfp . ucsf . edu/allOrfData . txt ) . Similarly , ambiguous localizations and localizations with <15 proteins were not used . The list of orthologous genes between the yeast and fly was provided by He and Zhang [46] , who used reciprocal best-hits in BLASTP searches to define gene orthology ( E-value cutoff = 10−10 ) . The same method was used to identify the yeast and nematode orthologous genes . The dN values between S . cerevisiae and S . bayanus orthologous genes were computed by a likelihood method and obtained from Zhang and He [45] . We used the parsimony principle to infer the PL ( i . e . , the number of gene loss events ) for each of the S . cerevisiae genes throughout the known phylogeny of 12 fungi . The protein sequences predicted from the complete genome sequences of the 12 species were downloaded from ftp://genome-ftp . stanford . edu/pub/yeast/data_download/sequence ( S . cerevisiae , S . bayanus , S . paradoxus , and S . mikatae ) , ftp://ftp . ncbi . nih . gov/genomes/Fungi ( Candida glabrata , Kluyveromyces lactis , Eremothecium gossypii , Debaryomyces hansenii , and Yarrowia lipolytica ) , http://www . broad . mit . edu/annotation/genome/neurospora/Home . html ( Neurospora crassa ) , http://www . broad . mit . edu/seq/YeastDuplication ( K . waltii ) , and http://www . sanger . ac . uk/Projects/S_pombe/ ( Schizosaccharomyces pombe ) . A S . cerevisiae gene is considered to be lost in species X if it does not hit any genes in X ( Evalue cutoff = 10−1 ) but has a hit in at least one species that is more distantly related to S . cerevisiae than X is related to S . cerevisiae . Here X refers to one of the ten fungi that are neither S . cerevisiae nor S . pombe , the latter being the most distantly related species to S . cerevisiae in our study . The growth rates of the yeast single-gene deletion strains were originally generated by the Stanford Genome Technological Center [72] , and we here used the dataset curated and provided by Zhang and He [45] . NMI was described in detail in [30] . Briefly , let us define the matrix N , where each row corresponds to a module in separation X and each column corresponds to a module in separation Y . Each member Nij in the matrix represents the number of nodes in the ith module of X that appear in the jth module of Y . The calculation of NMI is given by where nX and nY are the number of modules in module separation X and Y , respectively . The sum over row i of matrix Nij is denoted Ni , and the sum over column j is denoted Nj . If two module separations are identical , the NMI between them reaches the maximum value of 1 . Datasets used in this work and computer programs made for the analyses can be downloaded from http://www . umich . edu/~zhanglab/download . htm . | Many complex networks are naturally divided into communities or modules , where links within modules are much denser than those across modules . For example , human individuals belonging to the same ethnic groups interact more than those from different ethnic groups . Cellular functions are also organized in a highly modular manner , where each module is a discrete object composed of a group of tightly linked components and performs a relatively independent task . It is interesting to ask whether this modularity in cellular function arises from modularity in molecular interaction networks such as the transcriptional regulatory network and protein–protein interaction ( PPI ) network . We analyze the yeast PPI network and show that it is indeed significantly more modular than randomly rewired networks . However , we find little evidence that the structural modules correspond to functional units . We also fail to observe any evolutionary conservation among yeast , fly , and nematode PPI modules . We then show by computer simulation that modular structures can arise during network growth via a simple model of gene duplication , without natural selection for modularity . Thus , it appears that the structural modules in the PPI network may have originated as an evolutionary byproduct without much biological significance . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"evolutionary",
"biology",
"saccharomyces",
"drosophila",
"computational",
"biology"
] | 2007 | In Search of the Biological Significance of Modular Structures in Protein Networks |
Gene expression in a tissue-specific context depends on the combined efforts of epigenetic , transcriptional and post-transcriptional processes that lead to the production of specific proteins that are important determinants of cellular identity . Ribosomes are a central component of the protein biosynthesis machinery in cells; however , their regulatory roles in the translational control of gene expression in skeletal muscle remain to be defined . In a genetic screen to identify critical regulators of myogenesis , we identified a DEAD-Box RNA helicase , DDX27 , that is required for skeletal muscle growth and regeneration . We demonstrate that DDX27 regulates ribosomal RNA ( rRNA ) maturation , and thereby the ribosome biogenesis and the translation of specific transcripts during myogenesis . These findings provide insight into the translational regulation of gene expression in myogenesis and suggest novel functions for ribosomes in regulating gene expression in skeletal muscles .
Ribosome biogenesis is fundamental to all life forms and is the primary determinant of translational capacity of the cell . Ribosome biogenesis is a complex process that involves transcription , modification and processing of ribosomal RNA , production of ribosomal proteins and auxiliary factors and coordinated assembly of ribonucleoprotein complexes to produce functional ribosomes [1] . While previously considered a “house keeping” constitutive process , recent studies have shown that ribosome biogenesis is regulated differently between cells and can be modulated in a cell type-specific manner [2] . These differences are required to generate ribosomes of different heterogeneities and functionalities that contribute to the translational control of gene regulation by selecting mRNA subsets to be translated under specific growth conditions potentially by identifying specific recognition elements in the mRNA [3–5] . Protein synthesis is the end stage of the gene regulation hierarchy and despite the identification of translational regulators of specific genes , a systematic identification of translational regulatory processes critical for tissue specific control of gene expression is still lacking . Moreover , upstream regulatory factors/processes that regulate ribosome heterogeneity in a cellular and organ-specific context in vertebrates still remains to be identified . Skeletal muscle is a contractile , post-mitotic tissue that accounts for 30–50% of body mass . Skeletal muscle growth and repair is a highly coordinated process that is dependent on the proliferative expansion and differentiation of muscle stem cells that originate from the muscle precursor cells at the end of embryogenesis [6–8] . Activated muscle stem cells through symmetric/ asymmetric divisions give rise to daughter cells that maintain the muscle stem cell population ( satellite cells ) and committed myogenic progenitor cells ( MPC ) that fuse to form the differentiated skeletal muscle [6 , 9 , 10] . Defects in processes regulating muscle stem cells , myoblast fusion and differentiation therefore , constitute pathological pathways affecting the muscle growth and regeneration [11–17] . Although epigenetic and transcriptional controls of myogenesis have been studied extensively , the importance of translational regulation of these processes in skeletal muscle function is less defined . Transcriptional and translational analysis of myoblasts has shown that differential mRNA translation controls protein expression of specific subset of genes during myogenesis . Recent studies have revealed ribosomal changes during skeletal muscle growth and atrophy . An increase in ribosome biogenesis is often observed during skeletal muscle hypertrophy [18 , 19] . Ribosomal perturbations on the other hand are associated with skeletal muscle growth and diseases [20–23] . Deletion of ribosomal protein genes S6k1 and Rps6 in mice results in smaller myofibers and reduced muscle function . In several muscular dystrophy models and atrophied muscles , a reduction in ribosome number and/or activity is observed [22 , 24 , 25] . In spite of these dynamic changes observed in ribosomes during different growth conditions , our understanding of the mechanism ( s ) that regulate ribosomal biogenesis in skeletal muscle and eventually control the translational landscape during muscle growth is still poor . In a forward genetic screen to identify critical regulators of myogenesis in vivo , we recently identified a RNA helicase gene , ddx27 , that controls skeletal muscle growth and regeneration in zebrafish [26 , 27] . We further show that DDX27 is required for rRNA maturation and ribosome biogenesis in skeletal muscles . Strikingly , DDX27-deficient myoblasts exhibit impaired translation of mRNA transcripts that have been shown to control proliferation as well as differentiation of muscle progenitors during myogenesis . These studies highlight the specific role of upstream ribosome biogenesis processes in regulating tissue specific gene expression during myogenesis .
Zebrafish and human exhibit similar skeletal muscle structure and the molecular regulatory hierarchy of myogenesis is conserved between zebrafish and mammals [9 , 28–30] . Therefore , to identify regulators of skeletal muscle growth and disease in vivo , we performed an ENU mutagenesis screen in zebrafish [26] . Analysis of skeletal muscle of 4–5 dpf ( days post fertilization ) larvae of one mutant identified in the screen , Osoi ( Japanese for “slow” ) , displayed highly reduced birefringence in polarized light microscopy in comparison to the control , indicative of skeletal muscle defects ( Fig 1A ) . Positional mapping and sequencing of this mutation identified a 20 bp deletion in exon 18 of the ddx27 gene ( DEAD-box containing RNA helicase 27 ) ( Figs 1B , S1A and S1B ) . qRT-PCR analysis showed significantly lower levels of ddx27 transcripts reflecting probable nonsense-mediated decay ( S1C Fig ) . Overexpression of human DDX27 resulted in a rescue of skeletal muscle defects in mutant embryos , demonstrating functional evolutionary conservation and confirming that mutation in ddx27 are causal for the osoi phenotype ( Fig 1C ) . Evaluation of a large number of mutant embryos showed that homozygotes die by 6–7 dpf . qRT-PCR analysis of mouse Ddx27 mRNA demonstrated Ddx27 expression in several tissue types ( S1D Fig ) . Whole-mount immunofluorescence of wild-type zebrafish ( 4 dpf ) detected Ddx27 expression in Pax7 labeled muscle progenitor cells in skeletal muscle ( Fig 1D ) . To identify DDX27 expression domains in mammalian skeletal muscle , immunofluorescence was performed on single myofibers isolated from extensor digitorum longus ( EDL ) muscles of wild-type mice ( Fig 1E ) . Immunofluorescence on freshly isolated EDL myofibers ( Day 0 ) showed that Ddx27 is expressed in Pax7 positive satellite cells . EDL myofibers were cultured for 3 days , which results in activation ( day1 ) , proliferation and differentiation of satellite cells ( day 2 ) . Ddx27 expression remained high in activated satellite cells ( day1 ) . Ddx27 expression was observed in proliferating Pax7 positive nuclei as well as MyoD expressing nuclei . Subsequently , after culture of myofibers for 1–2 days , satellite cells typically undergo cell division . After 3 days in culture , proliferating satellite cells as well as proliferating myoblasts ( MyoD positive ) exhibited Ddx27 expression . To determine Ddx27 expression during muscle differentiation , western blot was performed on proliferating and differentiating C2C12 myoblasts ( Fig 1F ) . Ddx27 expression was detected in proliferating myoblasts and declined as cells are committed towards differentiation ( 100% confluence , day 0 ) . The downregulation of ddx27 corresponded to an increase in MyoG and myosin heavy chain expression ( MF20 ) . These data show that Ddx27 expression is high in proliferating satellite cells and myoblasts and is reduced in myoblasts as they are committed towards the terminal differentiation into myotubes . Finally , analysis of subnuclear localization by immunofluorescence revealed that DDX27 is co-localized with UBF ( fibrillar component ) and Fibrillarin ( dense fibrillary component ) in nucleoli of human myoblasts ( Fig 1G ) . No co-localization of ddx27 was detected with B23 , labeling the granular component of nucleoli . The nucleolus is the primary site of ribosome biogenesis , therefore , this sub-nucleolar organization of DDX27 suggests a functional requirement of DDX27 in rDNA transcription and/or pre-rRNA processing steps in skeletal muscles . To investigate the role of RNA helicase Ddx27 in skeletal muscle homeostasis , we analyzed the structure and function of mutant zebrafish embryos ( 2 dpf ) and larvae ( 4–5 dpf . Analysis of skeletal muscle histology at the end of primary myogenesis in 2 dpf embryos showed no visible differences between control and mutant skeletal muscles ( S2A Fig ) . Quantification of fiber cross section area ( CSA ) also showed no significant differences in control and mutant muscles at 2 dpf ( S2B Fig ) . The nuclear content of the myofibers was analyzed as fiber CSA per nuclei that was similar in control and mutant myofibers at 2 dpf , however showed a significant decrease in mutants at 4 dpf ( S2C Fig ) . This suggests that skeletal muscle development is normal during embryogenesis ( 0–2 dpf ) in ddx27 fish . During development , Ddx27 expression is observed during embryogenesis and persists during larval stages in zebrafish ( www . zfin . com ) . However , normal skeletal muscle growth during embryogenesis could be due to a functional redundancy with other family members , many of which have overlapping expression patterns with Ddx27 . Homozygous ddx27 mutant fish ( 5 dpf ) showed reduced fiber diameter ( Fig 2A and 2B arrow , inset ) and central nuclei as observed in several muscle diseases ( Fig 2A and 2B , arrowhead ) [31] . Electron microscopy of longitudinal sections of ddx27 mutant myofibers also showed disorganized myofibrillar structures in ddx27 mutant fish ( Fig 2C–2E , arrow ) . The mutant myofibers displayed smaller Z-lines and disorganized actin-myosin assemblies in comparison to controls suggesting differentiation defects of these muscles . Immunofluorescence of cultured myofibers from control and mutant zebrafish further showed a reduction and disorganization of skeletal muscle differentiation markers at 4 dpf , ( Actn2/3 and Ryr1 ) validating our earlier observation that absence of Ddx27 results in a defective differentiation of skeletal muscles ( S2B Fig ) . The centralized mutant nuclei were round in shape with highly enlarged nucleoli in comparison to controls ( Fig 2E , arrowhead ) . Cross-sections of muscles showed a reduction in myofiber diameter ( 52 ± 14% ) in ddx27 mutant fish in comparison to controls ( Fig 2F–2H ) . Skeletal muscles of mutant fish also displayed whorled membrane structures associated with nuclei ( Fig 2D , arrow ) . These whorled membrane structures are often observed in skeletal muscle of congenital myopathy patients , however their origin and biological significance remains unknown . Ddx27 belongs to a family of highly conserved RNA helicases in vertebrates . Therefore , to evaluate if the function of Ddx27 in skeletal muscle is conserved in vertebrates , myogenic differentiation of control and Ddx27 knockout C2C12 myoblasts was analyzed . C2C12 myoblasts exhibited reduced proliferation and impaired differentiation , that failed to form mature myofibers ( S2F and S2G Fig ) . These results show that Ddx27 deficiency affects skeletal muscle growth during zebrafish larval stages leading in to skeletal muscle hypotrophy and DDX27 functions in growth and differentiation of skeletal muscle are conserved among vertebrates . To identify molecular events leading to growth defects in ddx27 mutant zebrafish , qRT-PCR was performed during embryonic ( 2 dpf ) and larval ( 4 dpf ) stages . qRT-PCR revealed no significant expression changes in muscle progenitor cell markers , pax3 and pax7 during the embryonic stage ( Fig 2I ) . In zebrafish , the PAX7 orthologue is duplicated in two copies; pax7a and pax7b . pax7a expressing cells act to initiate myofiber formation post-injury , whereas pax7a/b expressing cells are required for myofiber growth [32] . Significant down regulation of both pax7a and b was observed during the larval stage in ddx27 mutant zebrafish . Concurrently , with a downregulation of pax3 and pax7a/b , we also observed a high expression of myoD , myf5 and desmin in mutant muscles suggesting a premature activation of the myogenic program in Ddx27 deficiency ( Fig 2I ) . However , this precocious activation of the myogenic program appears to be unable to progress to a proper differentiated state , as documented by low levels of late differentiation markers myog and mylz leading to disorganized sarcomeres . ddx27 zebrafish larvae exhibit significantly slower swimming than wild-type controls at 5 dpf ( 91 ± 29 mm/10 min for ddx27 vs . 16823 ± 214 mm/10 min for controls ) . To quantify the functional deficits in skeletal muscles of homozygous ddx27 mutant larvae ( 5 dpf ) , peak twitch and tetanic forces of individual zebrafish skeletal muscle preparations were measured following electrical stimulation . Substantially depressed twitch and tetanic forces as well as the slower rise and fall of tension were observed in ddx27 mutants ( Fig 3A ) . Absolute tetanic force was significantly less for ddx27 larvae compared to controls ( mean ± SD: 0 . 28 ± 0 . 18 vs . 1 . 43 ± 0 . 27 mN; p < 0 . 0001 ) ( Fig 3B ) . These force deficits in ddx27 persisted in the force measurements normalized for cross-sectional areas ( Fig 3C: 11 ± 8 vs . 48 ± 6 kPa; p < 0 . 0001 ) , suggesting that the depressed tetanic force of the ddx27 mutant preparations is primarily due to intrinsic skeletal muscle deficits . Mutant ddx27 preparations also showed significant reductions in mean absolute twitch force ( Fig 3D: 0 . 16 ± 0 . 12 vs . 1 . 11 ± 0 . 23 mN; p < 0 . 0001 ) and twitch force normalized to the CSA of individual larvae ( Fig 3E: 7 ± 5 vs . 37 ± 4 kPa; p < 0 . 0001 ) . Interestingly , twitch force was depressed to a relatively greater extent than tetanic force as revealed by the significantly reduced twitch to tetanic force ratios of the ddx27 preparations ( Fig 3F: 0 . 57 ± 0 . 15 vs . 0 . 78 ± 0 . 06 kPa; p = 0 . 0006 ) . In addition to these differences in force magnitude , the kinetics of force development and relaxation were also affected in mutant skeletal muscles . The ddx27 larvae displayed a significantly slower rise in the maximal rate of twitch tension development , +dP/dt ( Fig 3G: 1 . 06 ± 0 . 71 vs . 8 . 00 ± 1 . 30 kPa/ms; p < 0 . 0001 ) , and a significantly slower maximal rate of twitch tension relaxation , -dP/dt ( Fig 3H: -0 . 47 ± 0 . 24 vs . -2 . 55 ± 0 . 35 kPa/ms; p < 0 . 0001 ) . This disproportionate reduction in twitch vs . tetanic force and the slowing of twitch kinetics indicate a reduction in the quality of the contractile performance of the ddx27 larvae that are consistent with reduced motility of mutant muscles . To understand if the decreased Pax7 expression observed in ddx27 mutant is due to reduced Pax7 expression in MPCs or is caused by reduced number of MPCs , we performed whole mount immunofluorescence with Pax7 antibody during embryonic and larval stages . In zebrafish , Pax7 labels proliferative cells in dermomyotome , quiescent muscle stem cells and myoblasts that are required for muscle growth and regeneration [8 , 33] . Quantification of Pax7 positive cells by whole mount immunofluorescence showed no differences in control and mutant muscles during embryogenesis ( 2 dpf ) . However , a significant reduction in Pax7 expressing nuclei ( 40 ± 8 . 2% ) was observed in larval mutant skeletal muscle compared with control fish ( 4 dpf ) ( Fig 4A ) . This implies that reduced pax7 expression observed in ddx27 mutant fish is a consequence of reduced number of Pax7 cells in these fish . This decrease in Pax7 cells in mutants could either be due to a defect in proliferation of MPC population or an increased apoptosis in Ddx27 deficiency . To study the proliferative potential of Ddx27-deficient MPC , we performed 5-ethynyl-2’-deoxyuridine ( EdU ) incorporation assay and Pax7 immunofluorescence in control and mutant fish . The proportion of nuclei that were double labeled with Pax7+Edu+ in total Pax7 labeled population , decreased significantly to ~14 . 1% in ddx27 skeletal muscle in comparison to the control ( ~22 . 0% ) . This indicates that the proliferation of Pax7 cells is significantly reduced in ddx27 mutant fish ( Fig 4B ) . We also investigated if Ddx27 deficiency leads to increased apoptosis of Pax7 MPC population . Whole mount TUNEL staining and western blot analysis with caspase 3 antibody at 3 and 4 dpf in control and mutant fish did not reveal any significant increase in apoptotic changes in mutant muscles compared to controls at these stages suggesting that decreased proliferation associated with premature differentiation rather than enhanced cell death underlie reduced MPC population observed in mutant muscles ( S3 Fig ) . A stem cell niche equivalent to mammalian satellite cell system exists in zebrafish that involves migration and asymmetric division and/or proliferation of Pax7 expressing muscle progenitor cells to repair muscle upon injury [8 , 34 , 35] . Moreover , processes and timing of skeletal muscle repair in larval zebrafish are highly similar to adult mammalian skeletal muscle . As Pax7 MPC population also contributes to skeletal muscle repair in zebrafish , we next investigated if Ddx27 deficiency affects muscle repair in mutant fish . To induce muscle injury , cardiotoxin was injected into the epiaxial myotome of somites of larval fish at 3 dpf and skeletal muscles were analyzed at 5 dpf as described previously [9] . Whole mount immunofluorescence showed that cardiotoxin administration resulted in accumulation of a pool of Pax7 expressing cells at the site of injury in control fish which exhibited efficient myofiber repair post injury as seen by newly formed lighter stained actin myofibers ( Fig 4C ) . In contrast , injured Ddx27-deficient zebrafish muscles exhibited numerous degenerating myofibers and no accumulation of Pax7 expressing cells or muscle repair was observed post injury suggesting an impaired regeneration ( Fig 4C ) . These findings suggest that Ddx27 plays a pivotal role in skeletal muscle regeneration . The major steps in skeletal muscle repair in zebrafish involve proliferation of MPC population , migration to the injury site and fusion with the damaged myofibers or form new myofibers [9] . ddx27 mutant fish exhibit reduced MPCs proliferation as well as differentiation defects to form mature myofibers suggesting that either or both of these processes could underlie the repair defects observed in Ddx27 deficiency in skeletal muscle . Ddx27 is localized in the nucleolus that is primarily the site of ribosome biogenesis . Therefore , we evaluated nucleolar structure and functions to understand the impact of Ddx27 on these processes in skeletal muscle . Immunofluorescence with different nucleolar markers showed changes in localization of the fibrillary component marker Ubf ( labeling rRNA transcription sites ) from small punctate foci to larger condensed areas , suggesting a perturbation in active transcription sites in the nucleolus . Similarly , fibrillarin-enriched dense fibrillary component areas of early rRNA-processing regions were also disrupted and merged , forming larger , more condensed structures . Lastly , granular component of the nucleolus which is the site of late rRNA processing was also perturbed in the ddx27 mutant . B23 , a nucleolar granular component marker exhibited altered localization to nucleoplasm in mutant nucleoli in comparison to the nucleolar restricted expression controls suggesting a structural disruption of fibrillary and dense fibrillary component potentially disrupts the organization of granular compartment in mutant nuclei ( Fig 5A ) . These results suggest that Ddx27 deficiency disrupts sites of rRNA synthesis , processing and early ribosomal assembly . To investigate the effect of Ddx27 on rRNA synthesis , we performed in situ rRNA transcription analysis in zebrafish skeletal muscles ( 5 dpf ) by 5-EU labeling of newly synthesized rRNA . Following 5-EU labeling , immunofluorescence was performed to visualize MPCs ( Pax7 positive cells ) or myonuclei ( Actn2/3 labeling ) . Analysis of 5-EU signal in MPC population revealed a significant reduction in rRNA transcripts in Pax7 positive cells in ddx27 fish ( 30 ± 8% ) . Interestingly , a significant decrease in rRNA synthesis was also observed in myonuclei in control myofibers . Further , examination of mutant myonuclei revealed a reduction in rRNA synthesis in comparison to control myonuclei . Together , these results suggest that Ddx27 deficiency impairs rRNA synthesis directly in MPCs . As these MPCs differentiation to form myofibers , rRNA synthesis defect persists in myonuclei contributing to impaired ribosome biogenesis and reduced muscle function in a non-autonomous manner in myofibers ( Fig 5B ) . Next , to evaluate the role of Ddx27 on pre-rRNA processing , the pre-rRNA maturation pattern was evaluated in the skeletal muscle of ddx27 mutant zebrafish . Skeletal muscles ( containing MPCs and myonuclei ) were dissected from control and mutant larval fish ( 5 dpf ) and total RNA was isolated . Northern blotting was performed with different probes that were specific to various precursor rRNAs representing different pre-rRNA processing steps ( Fig 5C and 5D ) . ddx27 mutant skeletal muscle displayed significant accumulation of long pre-rRNAs ( precursors A and B ) , corresponding to 47S and 43S pre-rRNAs in the human rRNA processing pathway . Precursors C ( corresponding to human 32S pre-rRNA ) also accumulated , while precursors to the 18S rRNA ( precursors D , E ) decreased . This accumulation of early precursors and of precursors C suggests an early defect in the rRNA maturation process and indicative of delayed cleavages in the 5’ETS and 3’ETS . These results are in accordance with recent data showing that depletion of DDX27 in human cells leads to the release of an extended form of the 47S primary transcript [36] . In addition , our data reveal an accumulation of 41S pre-rRNAs and a concomitant decrease of 30S pre-rRNAs , which are indicative of an impaired cleavage at site 2 ( Fig 5C ) . To study the impact of rRNA maturation defects on ribosomes , we performed ribosomal profiling in control and mutant zebrafish muscle that revealed a significant decrease in free 60S large ribosomal subunits . A reduction of mature 80S monosomes as well as polysomes was also observed in mutant muscles ( Fig 5E ) . Together these studies show that ddx27 expression is required for the formation of mature rRNA species and thereby for biogenesis of functional polysomes in skeletal muscle . We next sought to investigate if the ribosomal deficits due to Ddx27 deficiency affect translation of global processes or a specific mRNA repertoire in skeletal muscle . To identify mRNA repertoire exhibiting perturbed translation in Ddx27 deficiency , polysome profiling and subsequent RNA sequencing was performed in Ddx27 knockout C2C12 myoblasts that exhibit proliferation and differentiation defects similar to zebrafish . Also , Ddx27 is highly expressed in proliferating C2C12 myoblasts suggesting that mRNA species identified from polysomal profiling will potentially be a direct consequence of Ddx27 deficiency . Polysomes ( actively translating ribosomes ) were purified from control and knockout Ddx27 C2C12 myoblasts grown in the proliferation media . RNA-sequencing of total and polysome bound mRNA transcripts revealed that 124 transcripts that showed increased and 300 which showed decreased association with polysomes were common in both control and mutant . 1057 mRNA transcripts were exclusively enriched in control polysomes whereas Ddx27 deficient polysomes showed an enrichment of 286 mRNA transcripts ( Fig 6 ) . Data analysis ( S4 and S5 Figs ) showed that control polysomes associated transcripts exhibited an enrichment in mRNAs encoding ribosomal , RNA polymerase , RNA degradation and splicing pathways suggesting a high requirement of these RNA metabolic processes during muscle cell growth . On the other hand , in DDX27 deficiency an enrichment of apoptotic and inflammatory pathway genes was observed suggesting that absence of DDX27 activates the atrophic processes in muscle . Interestingly , mutant fish did not exhibit any visible increase in apoptosis . The enrichment of apoptotic transcripts in mutants could potentially be due to an initiation of end stage changes as mutants die by 5 dpf . In addition , mRNAs required for protein biosynthesis ( amino acid biosynthesis , aminoacyl-tRNA biosynthesis ) showed significantly lower enrichment in mutant polysomes . These result suggest that DDX27 is crucial for the translation of mRNAs that are necessary for generating building blocks for active biosynthesis of proteins and suppression of transcripts associated with atrophic processes during muscle growth . Interestingly , a number of signal transduction pathways associated with skeletal muscle growth and diseases are also perturbed in DDX27 deficiency . Mutant polysomes exhibited a reduced association with Fgfr1 transcripts . FGF-signaling pathway is crucial for skeletal muscle growth and muscle specific ablation of Fgfr1 impairs proliferation of muscle satellite cells [37] . Additionally , an increase in mRNAs encoding members of MAP Kinase pathway that is associated with precocious differentiation was observed in ddx27 mutant zebrafish muscles [38] . Lastly , we identified several novel mRNAs that were highly enriched in control polysomes but decreased in DDX27-deficienct polysomes ( e . g . Ctsw , Hddc2 , Tagln , Wfdc1 Sprr2h , and Vmn2r78 ) ( Fig 6C ) . Many of these genes are involved in the maintenance of proliferative state of human ES and ipS cell however , their role/s in myogenesis are not known [39] . To confirmed the validity of our polysomal profiling data and expression of these transcripts in skeletal muscle we analyzed the expression of HDDC2 in control and Ddx27 mutant myoblasts . Immunofluorescence and Western blotting revealed a significant downregulation of HDDC2 protein in Ddx27 mutant myoblasts suggesting that HDDC2 may be contributing to DDX27 mediated muscle stem cell defects in skeletal muscle ( Fig 6D ) . A reverse analysis also identified the downregulation of 4 novel transcripts in control myoblasts that were enriched in Ddx27 polysomes , including A330074K22Rik , Fhdc1 , Gpr153 , and Ston2 and future studies will be able to identify their functional roles in skeletal muscle . In sum , polysome profiling revealed that DDX27 is required for the translation of mRNAs regulating RNA metabolism and signaling pathway that are crucial for muscle satellite cell proliferation and differentiation . In addition , we identified novel genes that are required for cellular proliferation and future studies on in vivo function of these genes in skeletal muscle may help to understand novel processes regulating muscle growth .
DDX27 belongs to the DEAD-box family of RNA helicases which represent a large protein family with 43 members that catalyze the ATP-dependent unwinding of double stranded RNA and variously functions in remodeling structures of RNA or RNA/protein complexes , dissociating RNA/protein complexes , or RNA annealing [40] . Studies in yeast and cellular models have shown that several DDX family members regulate different steps of rRNA processing; however , in vivo functions of these RNA helicases in vertebrates are still mostly unknown . While this plethora of RNA helicases implies the potential for functional redundancy , it also raises the attractive possibility that RNA helicases might perform a generic , unifying function in neuromuscular system by regulating RNA metabolism . DDX5/p68 RNA helicase promotes the assembly of proteins required for transcription initiation complex and chromatin remodeling during skeletal muscle differentiation [41] . Overexpression of DDX5 also restores the skeletal muscle function in a mouse model of myotonic dystrophy [42] . RNA helicases ( DDX1 and DDX3 ) play significant roles in muscle diseases by interacting with muscle specific transcription factors or with disease causing genes [43 , 44] . We further demonstrate that DDX27 is a nucleolar protein that is highly expressed in muscle progenitor cells . Although nucleolar proteins often have ubiquitous localization , high expression of DDX27 in satellite cells and myoblasts suggests a specialized role for this protein in controlling MPC-regulated processes in skeletal muscle . Ddx27 deficiency impairs skeletal muscle growth and regeneration in larval Ddx27-deficient zebrafish . The expression of pax7 RNA as well as number of Pax7 expressing MPC were found to be significantly reduced in skeletal muscles of ddx27 fish . This reduced number of Pax7 positive MPC population could be either due to a direct role of Ddx27 in regulating proliferation , premature differentiation or increased cell death . Our studies demonstrate that Ddx27 deficiency leads to a decrease in proliferation of the Pax7-positive muscle progenitor cell population in mutant skeletal muscles by accumulation of cells at the G1 stage of cell cycle ( S2H Fig ) . Considering Ddx27 is expressed in murine Pax7 positive cells and proliferating myoblasts ( Fig 1 ) , this reduced proliferation of MPC in zebrafish is likely due to a cell autonomous defect . Moreover , no significant increase in apoptosis was observed in ddx27 mutant muscles suggesting that the reduction in MPC is not a consequence of an increased cell death in mutants . During myoblast differentiation to myotubes , downregulation of Ddx27 expression is associated with an upregulation of MyoD , MyoG and subsequently , myosin heavy chain expression . These data suggest that Ddx27 expression may be required for maintaining the proliferative state of muscle progenitor cells in an autonomous manner whereas preventing their differentiation to mature muscles under normal conditions in a non-autonomous manner . This is supported by the observation that an absence of Ddx27 results in upregulation of myoD and myf5 in ddx27 mutants implicating a precocious differentiation of mutant myoblasts . Previous studies have shown that overexpression of MyoD is sufficient to induce myogenic differentiation suggesting a similar mechanism may be contributing to premature differentiation of Ddx27 deficient MPCs [45 , 46] . A lack of muscle repair in mutant larval fish further demonstrated that Ddx27 is also crucial for skeletal muscle regeneration . Skeletal muscle regeneration is a complex process involving migration and proliferation of muscle progenitor cells to the injury site followed by differentiation to form new myofibers or repair the existing damaged myofibers . In vivo imaging studies have shown that the process of muscle repair in larval zebrafish is highly similar to that in adult mammalian muscle . In zebrafish , Pax7-marked muscle progenitor cells migrate to the injury site , divide and undergo terminal differentiation and regenerate muscle fibers [9 , 32] . Our studies demonstrate an absence of Pax7 expressing MPCs at the injury site in ddx27 skeletal muscle . This could be either due to a defect in the proliferation of Pax7 muscle progenitor cells as observed by reduced Edu labeling of Pax7 positive nuclei in ddx27 mutant fish or due to the inability of muscle progenitor cells to migrate to the site of injury and/or fuse with damage fibers . Considering we observed reduced MPCs proliferation as well as a lack of proper myofiber differentiation in Ddx27 deficiency , either or a combination of both of these processes could be contributing to skeletal muscle repair defects in mutant fish . Follow up studies with transgenic reporter lines will help us investigate the contribution of different cell lineages and processes contributing to impaired skeletal muscle regeneration in Ddx27-deficient skeletal muscle . The nucleolus is a prominent organelle , central to gene expression , in which ribosome synthesis is initiated . Alterations in nucleolar structure are indicators of changes in cellular growth and proliferation , cell cycle regulation and senescence; therefore , identification of mechanisms that guide the nucleolar structure-function relationship in vivo is needed . Our studies show that Ddx27 is highly expressed in MPCs that regulate skeletal muscle growth and repair and Ddx27 deficiency results in rRNA synthesis defects in the MPC population . A number of recent studies have shown that nucleoli are actively involved in stem cell maintenance [47 , 48] . Actively proliferating stem cells and progenitor cells possess large nucleoli that change to smaller foci during differentiation . These changes in nucleolar structure are associated with reduced rDNA transcription and ribosome biogenesis during differentiation . Depletion of a number of nucleolar proteins results in reduced cell proliferation , abnormal cell cycle and enhanced differentiation , demonstrating that proper nucleolar function is required for self-renewal of ES cells [49 , 50] . During differentiation of mesenchymal progenitors in to osteoblasts , myoblasts or adipocytes , phenotypic regulatory factors critical for each lineage ( e . g . Runx2 , MyoD , Mgn or C/EBP ) suppress rDNA transcription suggesting regulation of rRNA synthesis by cell fate-determining factors is a broadly used mechanism for coordinating cell growth with lineage progression [51] . Notably , we identified that lack of Ddx27 also results in rRNA synthesis defects in myonuclei . As mutant MPCs form mature muscles , rRNA defects are also persisted in differentiated muscle fibers . There are 43 DDX proteins in vertebrates , several of which are also expressed in myonuclei ( DDX1 , 3 and 5 ) . Therefore , rRNA synthesis defects in ddx27 mutant myonuclei imply highly specialized roles of Ddx27 in skeletal muscle that are not compensated by other DDX family members . Collectively , these studies suggest that Ddx27 deficiency results in impaired proliferation and differentiation of MPCs leading to defective myofibers . This lack of Ddx27 in MPCs leads to a direct ribosomal defects in MPCs and have an indirect effect on myofibers derived from these MPC population . Molecularly , we established that Ddx27 regulates rRNA maturation and ribosome biogenesis . DDX27 is highly conserved in evolution , and mutation of the yeast ddx27 ortholog , drs1p , results in 25S rRNA maturation and 60S ribosome subunit biogenesis defects [52] . In zebrafish and mammalian cells , we identified an accumulation of primary transcripts and long-pre-rRNAs reflecting early rRNA maturation defects resulting in a decrease in functional ribosomes . Previous studies have shown that ribosome numbers are increased during skeletal muscle hypertrophy and reduced in atrophied muscles or skeletal muscle diseases [20–22] . However , the regulators of these processes remain to be identified . Our studies demonstrate that DDX27 is required for normal muscle growth by controlling rRNA maturation and ribosomal biogenesis . A number of recent studies have shown that ribosomes of different heterogeneities and functionalities exist and contribute to the translational control of gene regulation by selecting mRNA subsets to be translated under specific growth conditions [5 , 53] . For example , analysis of ribosome populations in mouse embryonic stem cells revealed that RPL10 enriched ribosomes preferentially regulate mRNAs controlling cellular growth whereas RPL10 depleted ribosomes exhibited increased binding to mRNA pools regulating stress responses and cell death [54] . Therefore , reduced proliferation of MPCs and terminal differentiation into mature myofibers in zebrafish could potentially be due defects in ribosome biogenesis affecting either the global translation rates or translation of specific subsets of mRNA required for proliferation and differentiation of MPC population . Polysomal profiling identified reduced association of mRNAs in FGF and MAPK signaling pathways that are known to regulate different stages of skeletal muscle growth in both cell autonomous and non-autonomous manners . These signaling pathways are associated with activation and myogenic commitment of muscle stem cells and thus could be contributing to the skeletal muscle defects observed in Ddx27 deficiency in zebrafish [37 , 38] . Polysomal profiling also revealed an enrichment of transcripts regulating RNA metabolism pathways in control polysomes that was lacking in the mutant muscles . Defects in RNA metabolism underlie disease pathophysiology in a number of neuromuscular diseases [55] . Therefore , identification of critical regulators of RNA based processes that contribute towards myogenesis is essential in order to understand the molecular basis of pathological changes in disease conditions . Interestingly , many pathways altered in mutant myoblast are also crucial for myofiber hypertrophy . Identification of these different processes in control and mutant muscles signifies an additional regulatory layer of translational regulation controlled by DDX27 that fine tunes the crucial processes associated with skeletal muscle growth . Interestingly , most of the translational changes observed in both control and mutant skeletal muscle are associated with genes that were previously not known to play significant roles in myogenesis . In particular , dysregulation of genes associated with cellular pluripotency and membrane remodeling suggests novel roles for these factors in skeletal muscle biology . Future work targeting these genes will help to illuminate additional processes that are crucial for myogenesis . This work provides new insight into nucleolar function in skeletal muscle growth , and opens a new avenue to explore the specific roles of nucleolar proteins and ribosome biogenesis in normal and disease muscle .
Fish were bred and maintained using standard methods as described ( 56 ) . All procedures were approved by the Boston Children’s Hospital Animal Care and Use Committee . Wild-type embryos were obtained from Oregon AB line and were staged by hours ( h ) or days ( d ) post fertilization at 28 . 5°C . Zebrafish embryonic ( 0–2 days post fertilization ) and larval stages ( 3–5 dpf ) have been defined as described previously [56] . Cardiotoxin-induced muscle regeneration studies in zebrafish were performed following previously published protocols [57] . Briefly , control and ddx27 mutant larvae ( 3 dpf ) were anesthetized and immobilized by embedding into 3% low melting agarose . Cardiotoxin ( 10mM , 1 μl ) was injected into dorsal somite muscles , and fish ( 8–10 , in 4 independent experiments ) were analyzed at 5 dpf by immunofluorescence analysis . Muscle degeneration and regeneration in mice was performed as described previously [58] . Functional experiments were performed as previously described [59] . Briefly , fish were studied in a bicarbonate buffer of the following composition: ( in mM ) 117 . 2 NaCl , 4 . 7 KCl , 1 . 2 MgCl2 , 1 . 2 KH2PO4 , 2 . 5 CaCl2 , 25 . 2 NaHCO3 , 11 . 1 glucose ( Dou et al . , 2008 ) . 4–5 dpf larvae were anesthetized in fish buffer containing 0 . 02% tricaine and decapitated . The head tissue was used for genotyping . The larval body was transferred to a small chamber containing fish buffer equilibrated with 95% O2 , 5% CO2 and maintained at 25°C . The larval body was attached to an isometric force transducer ( Aurora Scientific , Aurora , Ontario , CAN , model 403A ) and position motor ( Aurora Scientific model 308B ) using a 10–0 monofilament tie placed at the gastrointestinal opening and another tie attached several myotomes proximal from the tip of the tail . Twitches ( 200 μs pulse duration ) and tetani ( 300 Hz ) were elicited using supramaximal current delivered to platinum electrodes flanking the preparation . All data were collected at the optimal preparation length ( Lo ) for tetanic force . At the conclusion of the experiment , images of the preparations width and depth at Lo were obtained by carefully rotating the preparation about the gastrointestinal opening attachment point . Each image was analyzed by ImageJ using an internal length calibration . Preparation cross-sectional area ( CSA ) was calculated from width and depth measurements assuming the preparations cross-section was elliptical . Forces were calculated as active force , i . e . peak force minus the unstimulated baseline force , and are presented in absolute terms ( valid because all larvae were attached at a consistent anatomical landmark , the gastrointestinal opening ) as well as normalized to preparation CSA . The maximal rate of twitch tension development was determined as the maximal derivative of the force by time response between the onset of contraction and peak force . Likewise , the maximal rate of twitch tension relaxation was calculated as the first derivative of the force by time relaxation response , ranging from peak force until force had declined to approximately baseline . Statistical differences in the ddx27 ( n = 12 ) and control ( n = 10 ) group means were evaluated by a two-sample t-test . Whole mount immunofluorescence in zebrafish was performed as described previously [60] . For immunofluorescence studies in zebrafish myofiber culture , previously published protocol was followed and 30–40 myofibers were analyzed in each condition [61] . Paraformaldehyde ( 4% ) or methanol ( 100% ) were used as fixative for different antibodies . EDL culture and immunofluorescence was performed using previously published studies [62] . Primary antibodies used in this study were: α-actinin ( 1:100; Sigma , A7811 ) , RYR1 ( 1:100; Sigma , R-129 ) , Fibrillarin ( 1:50; Santa Cruz , sc-25397 ) , B23 ( 1:50; Santa Cruz , sc-5564 ) , UBF ( Sigma , 1:50 , HPA006385 ) , DDX27 ( 1:50 , Santa Cruz , sc-81074 ) , Pax7 ( 1:20; Developmental Studies Hybridoma Bank ) , Mef2 ( 1:20 , Santa Cruz , sc-17785 ) . For phalloidin staining , paraformaldehyde fixed embryos/larvae were incubated with phalloidin ( 1:40 , Thermo Fisher Scientific , A12379 ) ( and with primary antibody; for double immunofluorescence ) , overnight at 4°C followed by incubation with secondary antibody ( if primary antibody was used ) . Nuclear staining was done using DAPI ( Biolegend , 422801 ) . Secondary antibodies ( Thermo Fisher Scientific ) were used between 1:100–1:250 dilutions . Mouse C2c12 cells were cultured in growth medium consisting of DMEM supplemented with 20% fetal bovine serum . To induce differentiation , growth medium was replaced with the differentiation medium consisting of DMEM supplemented with 2% horse serum . Cells were maintained in the differentiation medium for 5 days . Western blot analysis was performed with Ddx27 ( 1:100 , Santa Cruz , sc-81074 ) , MyoD ( 1:200 , Santa Cruz , sc-760 ) , MyoG ( 1:200 , DHSB , F5D ) and MF20 ( 1:50 , DHSB ) . rRNA transcription was detected using the Click-iT RNA Alexa Fluor imaging kit ( C10329 , Invitrogen ) as described previously [63] . Briefly , to detect the synthesis of rRNAs in the nucleoplasm , control or ddx27 knockout zebrafish ( 10–12 ) or cultured myofibers ( 5 dpf ) were treated with 1μg/mL actinomycin for 20 minutes and then incubated for 2 hours with 1mM 5-ethyl uridine ( 5-EU ) in the presence or absence of actinomycin , fixed with paraformaldehyde ( 4% ) and incubated with Click-iT reaction cocktail for 1 hour . This was followed by immunofluorescence with Pax7 in zebrafish ( 1:20 , DHSB ) to detect MPCs and α-actinin ( 1:250 , Sigma: A7811 ) in cultured fish myofibers to label myonuclei . DNA was counterstained with DAPI . For the analysis of rRNA levels , the average 5-EU signal intensity in the nucleoplasm was measured from 20 nuclei per fish ( 10–12 fish in each group ) and three independent repeats using the ImageJ program . The level of rRNA was determined as the 5-EU dye signal level in actinomycin-treated samples subtracted from that of untreated samples , thus representing mature rRNAs in control and mutant fish or myofibers as described previously [63] . In order to analyze the precursors to the 28S and 18S rRNAs , total RNA samples were isolated from skeletal muscle of control and mutant larvae ( n = 100 ) and separated on a 1 . 2% agarose gel containing 1 . 2% formaldehyde and 1× Tri/Tri buffer ( 30 mM triethanolamine , 30 mM tricine , pH 7 . 9 ) . RNAs were transferred to Hybond N+ nylon membrane ( GE Healthcare , Orsay , France ) and cross-linked under UV light . Membrane hybridization with radiolabeled oligonucleotide probes was performed as described ( Preti , 2013 ) . Signals were acquired with a Typhoon Trio PhosphorImager and quantified using the MultiGauge software . The human probes were: 5’-TTTACTTCCTCTAGATAGTCAAGTTCGACC-3’ ( 18S ) , 5’-CCTCGCCCTCCGGGCTCCGTTAATGATC-3’ ( 5’ITS1 ) , a mixture of 5'-CTGCGAGGGAACCCCCAGCCGCGCA-3' ( ITS2-1 ) and 5'-GCGCGACGGCGGACGACACCGCGGCGTC-3' ( ITS2-2 ) , 5’-CCCGTTCCCTTGGCTGTGGTTTCGCTAGATA-3’ ( 28S ) , 5’-GCACGCGCGCGCGGACAAACCCTTG-3’ ( 28S-3’ETS ) . The zebrafish probes were: 5’-GAGGGAGGCGCGTCGACCTTCGCTGGGC-3’ ( 3’ETS ) , 5’- CAGCTTTGCAACCATACTCCCCCCGGAAC-3’ ( 18S ) , 5’-GAGATCCCCTCTCGAACCCGTAATGAT-3’ ( ITS1 ) , 5’-GAGCGCTGGCCTCGGAGATCGCTGGGTCGC-3’ ( ITS2 ) , 5’-CCTCTCGTACTGAGCAGGATTACTATTGC-3’ ( 28S ) . For ribosome profile analysis zebrafish ( 100 larvae ) were treated with cycloheximide ( Sigma , 100ug/ML ) for 10 minutes at room temperature . Subsequently , skeletal muscle was dissected and flash frozen in liquid nitrogen in lysis buffer ( 10mM Tris-Cl , pH 7 . 4 , 5mM MgCl2 , 100mM KCl , 1% TritonX-100 ) . To purify ribosomal fractions , cell lysate ( 0 . 75 ml ) was layered on 10–50% sucrose gradient and centrifuged at 36 , 000 rpm using SW-41 Ti rotor , 2 hours at 4°C . The fractions were collected and were analyzed at 254nm . Polysome ribosome fractions were prepared from control and Ddx27 C2C12 myoblasts . Equal number of control and Ddx27 knockout C2C12 myoblasts were plated in 10 cm dishes in the proliferation media and collected after 24 hours for polysomal analysis . Polysomes were isolated as described above for zebrafish muscle and fractions were pooled together , treated with proteinase K and RNA was isolated using acid phenol-chloroform extraction and ethanol precipitation . As Ddx27 myoblasts exhibit a reduced proliferation , equal amounts of proteins were used to fractionate polysomes from control and KO myoblasts . Deep sequencing libraries were generated and sequenced as described [64] . Ribosomal profiling was repeated in triplicate and principal component analysis was performed to identify variation between samples . The Spearman R2 was > 0 . 9 for all replicates ( except one sample ) that was subsequently removed from follow up analysis . Splice-Aware alignment program STAR was used to map sequencing reads to Mus musculus ( mm10 build ) . R package “edgeR” was employed to identify differential gene expression calls from these sequence reads . Gene expression was considered to be up-regulated if log2FC> +1 or downregulated if the log2FC< -1 ( FC = fold change of average CPM ) with respect to the condition being compared at a false discovery rate <0 . 05 . Functional ontological classification of different gene lists was performed by DAVID . To analyze the proliferating cells in C2C12 cell cultures and zebrafish embryos , Edu labeling was performed . For zebrafish , 20–25 embryos ( 48 hrs ) were placed in 1mL of 500μM EdU/10% DMSO in E3 media and incubated on ice for 2 h . Embryos were transferred back to the incubator at 28 . 5°C and samples were collected at desired time intervals and fixed in methanol ( -20°C , 20 mins ) . Fish were permeabilized with 1% tritonX-100/PBS for 1 hr and Click-iT reaction cocktail ( Thermo Fisher Scientific ) was added and incubated in dark for 1 hr at room temperature . Immunofluorescence was performed with Pax7 antibody ( 1:10 DHSB ) and analyzed ( 10–12 larvae ) by confocal microscopy . Proliferation was analyzed by calculating the proportion of Pax+Edu+ double positive cells in total Pax7 labeled population in control and mutant fish . For analyzing proliferating C2C12 cells , equal number of control and mutant cells ( to 40–50% confluency ) were plated for 12–14 hours . 2X EdU solution was added to the cells and incubated for 2 hrs at 37°C . After incubation , cells were permeabilized with 0 . 5% triton-100 and labeled with Click-iT reaction cocktail as described for zebrafish . Immunofluorescence was performed with Pax7 antibody ( DHSB ) and nuclei were stained with DAPI . Apoptosis was performed on 3-4dpf zebrafish larvae by in situ cell detection kit ( Roche ) or western blot analysis using caspase 3 antibody ( ab13847 , Abcam ) . Zebrafish swimming behavior was quantified by an infra-red tracking activity monitoring system ( DanioVision , Noldus , Leesburg , VA , USA ) . Control or ddx27 mutant larvae were placed individually into each well of a 24 well plate in dark for 10 minutes . The activity of these larvae was recorded during a follow-up light exposure of 20 minutes . Four independent blind trials were performed and mean velocity , total distance and cumulative duration of movement were recorded . Reported values reflect an average of 30–35 control or mutant larval fish . Quantification of myofiber size , area , western blots and northern blots were performed using the ImageJ program . Data were statistically analyzed by parametric Student t-test ( two tailed ) and were considered significant when P<0 . 05 . All data analyses were performed using XLSTAT software . Additional detailed experimental details are provided in the supplemental material . | Inherited skeletal muscle diseases are the most common form of genetic disorders with primary abnormalities in the structure and function of skeletal muscle resulting in the impaired locomotion in affected patients . A major hindrance to the development of effective therapies is a lack of understanding of biological processes that promote skeletal muscle growth . By performing a forward genetic screen in zebrafish we have identified mutation in a RNA helicase that leads to perturbations of ribosomal biogenesis pathway and impairs skeletal muscle growth and regeneration . Therefore , our studies have identified novel ribosome-based disease processes that may be therapeutic modulated to restore muscle function in skeletal muscle diseases . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] | [
"medicine",
"and",
"health",
"sciences",
"fish",
"skeletal",
"muscles",
"vertebrates",
"animals",
"animal",
"models",
"osteichthyes",
"muscle",
"regeneration",
"developmental",
"biology",
"model",
"organisms",
"organism",
"development",
"stem",
"cells",
"experimental",
"organism",
"systems",
"immunologic",
"techniques",
"morphogenesis",
"cellular",
"structures",
"and",
"organelles",
"research",
"and",
"analysis",
"methods",
"musculoskeletal",
"system",
"muscle",
"differentiation",
"animal",
"cells",
"immunoassays",
"muscles",
"myoblasts",
"immunofluorescence",
"ribosomes",
"biochemistry",
"zebrafish",
"rna",
"eukaryota",
"ribosomal",
"rna",
"anatomy",
"nucleic",
"acids",
"cell",
"biology",
"regeneration",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"non-coding",
"rna",
"organisms"
] | 2018 | RNA helicase, DDX27 regulates skeletal muscle growth and regeneration by modulation of translational processes |
Structural plasticity governs the long-term development of synaptic connections in the neocortex . While the underlying processes at the synapses are not fully understood , there is strong evidence that a process of random , independent formation and pruning of excitatory synapses can be ruled out . Instead , there must be some cooperation between the synaptic contacts connecting a single pre- and postsynaptic neuron pair . So far , the mechanism of cooperation is not known . Here we demonstrate that local correlation detection at the postsynaptic dendritic spine suffices to explain the synaptic cooperation effect , without assuming any hypothetical direct interaction pathway between the synaptic contacts . Candidate biomolecular mechanisms for dendritic correlation detection have been identified previously , as well as for structural plasticity based thereon . By analyzing and fitting of a simple model , we show that spike-timing correlation dependent structural plasticity , without additional mechanisms of cross-synapse interaction , can reproduce the experimentally observed distributions of numbers of synaptic contacts between pairs of neurons in the neocortex . Furthermore , the model yields a first explanation for the existence of both transient and persistent dendritic spines and allows to make predictions for future experiments .
The structure of neocortical networks of neurons changes in time: new synapses are formed , maturate , and eventually are pruned again , in the adult as well as in the developing animal [1] , [2] , for recent reviews see [3] , [4] , [5] . The majority ( about ) of excitatory synaptic contacts terminate on dendritic spines [6] , and dendritic spines almost always ( ) form a synapse [7] . The synapses on dendritic spines are highly dynamic [8] , [9] , for example [10] found an average spine turnover of in primary visual cortex and of in somatosensory cortex . Yet in the adult animal , the statistics of the numbers of synapses are preserved over time , indicating that synapse creation and pruning balance each other [11] , [12] , [13] . According to theoretical studies on associative networks , structural plasticity enhances the memory capacity of a network substantially [14] , [15] , and has been shown to be related to motor learning in the brain [16] . The three studies [17] , [18] , [19] reported the distributions of numbers of synaptic contacts for different intra-cortical synapses in rat somatosensory cortex . Fares et al . [20] subsequently analyzed whether the reported distributions could result from random and independent synaptic contact formation , given a set of potential sites ( close appositions ) between axons and dendrites of reconstructed cells . As they showed , independent formation of synaptic contacts alone cannot explain the distributions . In addition a cooperative pruning mechanism , by which synaptic contacts that constitute a single synapse stabilize each other , is required to explain the observed distributions . Here we build on grounds of this work and go beyond it in two aspects: Primarily , we consider synaptic processes that operate continuously in time . Secondly , we investigate an explicit candidate mechanism for the cooperation between synaptic contacts: Local correlation detection at the dendritic spines and thus dependent pruning and maturation of spines . Recently Kasai et al . [21] summarized known properties of the plasticity of dendritic spines . Their model [22] describes the dynamics of the volume of dendritic spines . Here we restrict this model to three distinct categories of synapse states and introduce an explicit spike-timing dependence . Other models of structural plasticity [23] , [24] , [25] are based on the firing rate of the neurons . Consequently , in these models spike-timing and correlations of the spiking activity do not play a role , so they cannot show the mechanism of synaptic cooperation that we hypothesize here . The relative timing of pre- and postsynaptic activity indeed influences structural plasticity at the dendritic spine [26] . In contrast to previous models , the model of Helias et al . [27] is sensitive to the spike-timing of the pre- and the postsynaptic cell and describes structural plasticity in biophysical terms of protein kinetics in response to synaptic input . Here we choose an intermediate scale by still describing single synaptic contacts , but with a higher level of abstraction than previous work [22] , [27] . The goal of the present work is to demonstrate the potential of local correlation detection at the spine , while making minimal assumptions about the involved biophysical processes . The assumptions entering our model , as introduced in detail in Methods , required to qualitatively explain the experimental results , are: a ) presynaptic release of glutamate causes postsynaptic depolarization at excitatory synapses , b ) depolarization electrically spreads within the dendrite , c ) there is a correlation sensing mechanism sensitive to the relative time of presynaptic and postsynaptic firing ( e . g . NMDA receptors ) that causes downstream effects on the evoked synaptic amplitude in a spike timing dependent plasticity ( STDP , [28] ) like manner , d ) synapses with small amplitude are more likely to be pruned than strong ones . Because of its analytical tractability , we can compute the steady state of our model and match its parameters to experimental reference data , analogous to Fares et al . [20] . Our results show that no direct signaling between synaptic contacts is necessary to explain cooperative synapse formation . In contrast , it suffices that distinct synaptic contacts cooperate in exciting the postsynaptic neuron , and thereby indirectly affect spike-timing dependent structural plasticity at other synaptic contacts .
Let us first introduce a simple model for the correlation detection at the postsynaptic dendritic spine . An action potential of the postsynaptic neuron causes a depolarization at the site of each dendritic spine . The spine has the biophysical substrate to maintain a signal that depends on the time of the action potential in relation to the time when a presynaptic impulse arrived [27] . Here , we call this signal the correlation trace and assume a phenomenological model: if the presynaptic neuron spiked shortly before the postsynaptic one , the correlation trace is increased by . We call this a causal event . For the opposite relative timing , called an anti-causal event , the trace is decreased by . The trace therefore counts causal and anti-causal combinations of pre- and postsynaptic spikes . Further we assume that the correlation trace is forgetful: it has a leak with the time constant , and we also assume that there is some additive noise in the process . The dynamics of the correlation trace at the synapse is given by the stochastic differential equation [29] ( 1 ) where is the spike train of the postsynaptic neuron , with the spike times , the factor specifies if the particular spike is counted as a causal or anti-causal event , and is an additional white noise with mean zero and infinitesimal variance . Mathematically , the trace is identical to a shot noise [30] with the exponential kernel driven by the stochastic input process . Let us now introduce a minimal model of correlated spiking of the neurons . For each postsynaptic spike , we speak of a causal event at a given synaptic contact if the closest spike of the presynaptic neuron occurred prior to the postsynaptic one ( because it could have caused the postsynaptic spike ) . If the closest presynaptic spike occurred after the postsynaptic one , the event is called anti-causal . Suppose that the probability for a causal event is given by . If both of the neurons fire independently , then . We define with probability and with probability for each postsynaptic spike in . Strictly speaking , the process defined by ( 1 ) is unphysical , since through it depends on events in the near future ( because the time of the next presynaptic spike has to be known ) . We hence consider ( 1 ) as an effective , adiabatic description of the correlation trace , since we are only interested in the statistics of the trace on long timescales . A process like ( 1 ) could result from several biophysical implementations that do in fact respect causality . For example the synaptic weights in phenomenological models of spike timing dependent plasticity , for which causal implementation are known [31] follow dynamics similar to ( 1 ) . An example of a cellular mechanism to implement ( 1 ) is the number of activated CaMKII macro-molecules [27] or long-term potentiation [32] , [33] , [34] . Now let us further assume that postsynaptic spikes occur according to a Poisson point process with rate . The firing rate comes about through integration of thousands of synaptic inputs , and the particular synaptic connection modeled here only provides a small contribution to . Since structural plasticity is known to be a slow process compared to the activity of neurons , and since the time constant of possible candidate mechanisms for the correlation trace can be considerable [35] , a large integration time constant is reasonable , such that . Then the equilibrium probability distribution of is a normal distribution with mean and variance , ( 2 ) ( 3 ) Eqs . ( 2 , 3 ) can be obtained by considering two independent stochastic processes and with and . Then and mean and variance of follow from summing the respective statistics of and , which can be obtained using standard techniques [30] . Note that is independent of . The probability of causal spike pairings depends on the number of active synapses connecting the presynaptic neuron to the postsynaptic one , because each excitatory synapse increases the chance of the presynaptic neuron to make the postsynaptic neuron fire . As demonstrated for integrate-and-fire neurons in [36] , the probability of a spiking response to a presynaptic spike is proportional to the synaptic weight of the input spike for a wide range of magnitudes of the synaptic strength . If the membrane potential of the postsynaptic neuron integrates the inputs linearly , the synaptic weight of the input from the presynaptic neuron is proportional to the number of active synaptic contacts between the neurons . So the probability of a spike response , and so of a causal event , rises proportionally to the number of active synaptic contacts . Effectively we thus assume , where is the number of active synaptic contacts between the presynaptic to the postsynaptic neuron , is the response probability per synapse , and . The two components of the probability can be interpreted as the probability for a causal event by chance due to the Poisson firing of the postsynaptic neuron with rate and the probability exceeding chance level triggered by the arrival of the presynaptic spike . In order to obtain an estimate of consider a single synaptic contact between two neurons that , upon activation , causes an excitatory postsynaptic potential ( EPSP ) with amplitude . For a leaky integrate-and-fire model neuron in the asynchronous state [37] resembling cortical activity , we can read off the response probability of a neuron to such a voltage jump from Fig . 4D in [36] . So we have with . The respective values of the EPSP size per contact have been reported along with the reference datasets in [17] , [18] , [19] , and we list them among the model parameters in Tab . 1 . We thus arrive at the estimate for the probability of a correlated pairing ( 4 ) and thus with ( 2 ) the mean of the stationary distribution of is ( 5 ) Note that the linear model ( 4 ) for the probability of causal spike pairings may for large and yield values of , which are nonsensical . A consistent definition of should saturate when reaching the value . Taking into account this saturation at large , however , would not make a difference for the models considered here , because all solutions for synapse distributions found below exhibit vanishing probably throughout at such large values of . So far , we hypothesized a generic correlation detection mechanism at each synaptic contact and computed its equilibrium statistics ( 5 ) and ( 3 ) for multiple excitatory contacts between two neurons . In our model the values of the correlation trace of each synaptic contact follow a normal distribution , specified by the its mean and variance , which depend on the parameters . Note that only the mean depends on the number of active contacts , whereas is a constant . To reduce the amount of free parameters of the model , we further set which is a reasonable choice for neocortical neurons . The synaptic correlation trace can guide structural plasticity . Because of a lack of detailed knowledge about the biomolecular mechanisms involved [38] we again employ a simple effective model . As structural plasticity is a slow process compared to the spiking activity of neurons , we assume that the distribution of the correlation trace at each synaptic contact is effectively in stochastic equilibrium throughout . This is also known as an adiabatic approximation . Let us now assume that a structural change of the synaptic contact is initiated when the correlation trace crosses a boundary value . A biochemical mechanism underlying this assumption could be the activation of a signaling pathway when a specific number of activated CaMKII molecules is reached [39] . Given the random trajectory of the correlation trace , we need to know how long it takes until crosses the boundary upon which the synaptic contact makes the transition . This is known as a first passage time problem with an absorbing boundary . The inverse of the mean first passage time is called the escape rate . For simplicity , let us approximate as a Brownian motion with the same infinitesimal mean and variance as the actual process ( 1 ) . Then , according to the Arrhenius approximation [40] , the escape rate is ( 6 ) in the case that the values of are far from the boundary , such that . Here the proportionality constant is called the Arrhenius constant . Now what happens to the rate of structural changes when approaches ? If we take the model of a boundary crossing process seriously , then the escape rate should diverge as . However , in a biological system it is more plausible that the rate of structural changes converges to a certain maximum rate which the cellular machinery can achieve . Based on this argument we construct our model for the rate of structural changes by extrapolating from the Arrhenius approximation ( 6 ) , forcing it to eventually converge to a plateau , ( 7 ) with . Depending on the sign of , either approaches or departs from the plateau for increasing . As our assumptions about the biophysical implementation are quite general , they can model maturation , shrinkage and pruning of a synaptic contact alike . However , these are distinct processes that take place during different stages in the life cycle of a synaptic contact . For example , we cannot assume that the correlation detector noise has the same magnitude for small and large dendritic spines since the number of channels mediating the signal might be different for the two . Therefore we use the model ( 7 ) for maturation , shrinkage and pruning , but choose a different set of transition parameters and correlation trace noise for each case . We decorate quantities associated to maturation with , those associated to shrinkage with and pruning with . So the rate of maturation transitions is defined as ( 8 ) the rate of shrinkage transitions as ( 9 ) and the rate of pruning as ( 10 ) The correlation trace parameters are assumed to be identical for both thin spines ( inactive synaptic contacts ) and large spines ( active synaptic contacts ) , which will be defined below . The pruning rate uses the same parameters as shrinkage , except for the noise magnitude of maturation , because pruning is assumed to take place in thin spines . Apart from the activity dependent transitions , the model also includes intrinsic fluctuations as in [22] , see Fig . 1B . We assume that random maturation ( enlargement ) , shrinkage and pruning of a spine occurs constantly with the rate , and the creation of new thin spines with rate . Let us summarize the structural plasticity model we have introduced . At each of the synaptic contacts between a pair of neurons , a correlation trace is formed by counting causal and anti-causal pre-post spike pairings . The distribution of the values of the correlation trace depends on the number of active synaptic contacts since they all contribute to firing the postsynaptic neuron . We have further assumed that the activity-dependent structural changes of synaptic contacts depend on the correlation traces . Finally we also included intrinsic fluctuations of the synapse configuration . Above we defined a model for structural plasticity for the synapse between a pre- and a postsynaptic neuron . Although in this model the individual synaptic contacts may continuously change , the state of the synapse develops towards a stable steady state . A synapse typically consists of many individual synaptic contacts , as depicted in Fig . 1A . The neocortex is densely packed with only very limited unoccupied extracellular space . Accordingly , pairs of neurons cannot form arbitrary numbers of synaptic contacts [41] . Fares et al . [20] investigated reconstructed cortical tissue and counted the numbers of close appositions between pairs of neurons . At such a close apposition a synaptic contact may form , but is not necessarily present . Describing these results statistically , a probability distribution for the number of close appositions between two neurons can be obtained [20] . At each of the close appositions , the neurons may form a synaptic contact; in our model , we treat the different volumes of spines and EPSP amplitudes in a coarse-grained fashion , distinguishing just three different states for each contact to occupy – active , inactive , or unrealized . An active synaptic contact here describes a larger dendritic spine that contains both AMPA and NMDA receptors . An inactive contact models a thin , either newly formed or recently shrunk dendritic spine that has much less AMPA receptors [42] , [43] and contributes little to firing the postsynaptic neuron . An unrealized contact , finally , is a close apposition where no contact has formed , but might be formed in the future . It is a close apposition without an established synaptic contact and corresponds to the potential synapse in [20] . A similar model has previously been proposed in the context of associative networks [44] , [45] . We denote the numbers of synaptic contacts in these three states by , and respectively . Since at any time , the state of a synapse is unambiguously defined by the combination of the number of active and inactive contacts . Now consider an ensemble of independent synapses , each with the maximum contact number . The probability of a synapse to be in the state evolves in time according to the Master equation [46] ( 11 ) The first term sums up the rate of leaving the state by all possible transitions . The second and third terms sum up all possibilities to go into state from other states by shrinkage and pruning , and the fourth term by maturation . The last term considers the transitions due to the creation of inactive synapses , the rate of which is given by ( see also Fig . 1B ) . The steady state distribution does not depend on the time scale of the transition rates , so we can consider the constants in units of . The time scale of the structural plasticity is then set by . In ( 25 ) below we will see how the value of can be determined by experimental data . To determine the steady state configuration of the synapse , let us introduce a numbering of all the possible synaptic states , such that the probability of each state is represented by the value , with a one to one correspondence between indices and states . Then ( 11 ) can be written as ( 12 ) where the entries of the matrix can be read off the Master equation . Since describes a Markov process it is column-stochastic ( which means all columns sum up to zero ) . Since the process is irreducible , according to the Perron-Frobenius theorem there is only one stationary solution . We can determine the stationary probability distribution by solving under the constraint that . We implemented the construction of the matrix efficiently using Cython [47] and solved for the stationary solution using Scientific Python [48] . The stationary distribution depends on the number of close appositions . [20] have estimated the distribution for the three types of intra-cortical connections that we consider . We incorporate this by determining for each separately , and subsequently compute the averaged distribution ( 13 ) Fares et al . [20] provide the distribution for up to . For comparison with the reference datasets ( see below ) , we are merely interested in the marginal probability of a certain total number of synaptic contacts , disregarding whether they are active or inactive . The marginalization can be obtained from by summing over all states with , or more conveniently phrased as ( 14 ) where the function returns the value of of the state with index , and equals if is true and otherwise . Analogously , the marginal average distributions of the number of active and inactive synapses are ( 15 ) ( 16 ) In the experimental studies [17] , [18] , [19] a set of occurrence frequencies of numbers of synaptic contacts for several pairs of neurons was obtained , each for three different types of intra-cortical projections . Complementing this , for the same three datasets , the probability of a pair of neurons to be connected with at least one active contact can be estimated [20] . The distribution of the numbers of active synaptic contacts serve as reference data in our study . For each of the three datasets , we transform the reported data to the probability mass function ( 17 ) we evaluate ( 14 ) and obtain the residuals ( 18 ) for . The residuals are scaled by the maximum of the reference distribution to enable comparison of the quality of the fits across reference datasets . We minimize the sum of squared residuals ( 19 ) using the Levenberg–Marquardt algorithm , applying its implementation from Scientific Python [48] . We call the error of the model . The optimization problem has several local minima , so we initialize the optimization procedure at many points in the -dimensional parameter space and compare the values of to which the optimization converged . Specifically , we choose four different initial values in each parameter dimension , which makes a total of distinct optimization runs per reference dataset . The parameter sets which resulted in a minimal value of are shown in Tab . 1 , along with additional information on the model , and the resulting equilibrium distributions are shown in Fig . 2 . We also obtained parameter sets which yield good fits to two reference distributions simultaneously . The resulting distributions for the connections L4-L4 and L5-L5 are displayed in Fig . 3 . Here the fit error was defined as the sum of the errors ( 19 ) of both distributions , . With respect to this the same optimization procedure was performed . Here we compute the average lifetime of an inactive synaptic contact and of an active contact in the equilibrium state of the synapse model , see Fig . 1a for the possible transitions . We define the lifetime as the expected time until the contact is pruned . Consider an active contact in a synapse . It may become an inactive contact either through intrinsic or activity dependent shrinkage . The mean time up to the transition from active to inactive is . An inactive contact , on the other hand , might make a transition to the active state ( maturation ) , which would take the time , or to the unrealized state ( pruning ) in the time . The mean time until the first transition , either maturation or pruning , is . Either of the two transitions happens with a probability given by the fraction of rates involved , and analogously . If the inactive contact becomes active , then it will become inactive eventually , and subsequently might be pruned or become active again . Accounting for the possible paths the inactive contact may take upon its first transition we obtain the expected lifetime of the inactive contact as ( 20 ) where is the lifetime of an active contact in a synapse that has active contacts . In turn , starting from an active contact just adds one active to inactive transition , so ( 21 ) Inserting ( 21 ) and the definitions above into ( 20 ) yields ( 22 ) We average the lifetimes across the equilibrium probability distribution of synapse states and obtain ( 23 ) ( 24 ) To match the time scale of structural development of our model to what is known from in-vivo studies we compute the spine turnover ratio as it is defined in [10] , ( 25 ) Here and are the numbers of gained and lost spines during a given period of time , and is the number of spines observed . In our model , the expectation values of these quantities are given aswhere and are given in units of . So we obtain in units of from ( 25 ) . In rat somatosensory cortex [10] found . Accordingly ( 25 ) sets the time scale of the model to . In this section we consider a simplified version of our model which does not include inactive synaptic contacts ( thin spines ) . In that model , at a close apposition there can be either no synaptic contact or an active synaptic contact . Between these two states transitions are allowed just as between the inactive and the active state in the full model ( see Fig . 1B ) , but here we call them ( creation ) and ( pruning ) , which may yet be arbitrary functions . Assume there are close apposition between a pair of neurons . Then the state of the synapse is defined by the number of active connections . Let us denote the probability of the state by here . In stochastic equilibrium the probability fluxes into and out of the state must balance , so for it must hold thatfrom which follows that ( 26 ) Now consider the case of . In all three reference datasets , , as can be seen in Fig . 2a1–a3 . According to ( 26 ) this requires , which can only be achieved by or . In contrast , around the secondary peak of the reference distributions , ( 26 ) entails that and must be of comparable magnitude . More specifically , the right hand side of expression ( 26 ) has to change from values larger than to values smaller than as passes the secondary peak from below . To satisfy these requirements , even only approximately , demands highly non-monotonous choices of the functions and that are difficult to justify biophysically . Once the model includes the intermediate state of the inactive synaptic contact , however , it is possible to find biologically plausible parameter sets to explain the reference distributions , as described in the rest of the paper .
Kasai et al . [22] monitored the temporal evolution of the volume of dendritic spines and described it as a random walk with volume dependent drift and diffusion components . According to their findings , newly formed dendritic spines are small , and accumulate AMPA receptors as the spine volume increases . Thus a small spine can grow or disappear , and a large spine can shrink . Spines of all volumes , however , were found to contain NMDA receptors [42] . The study by Holtmaat et al . [10] suggests that thin spines are more readily pruned than thick spines , and that they may be of a lower efficacy or NMDA receptor-only ( inactive ) synapses . It was previously suggested 43 , 49 that small spines might correspond to silent synaptic contacts . In our model , we distinguish between only three states that each synaptic contact can occupy: active , inactive and unrealized , without considering the spine volume and channel density of the dendritic spines explicitly . These states correspond , respectively , to large spines , thin spines and close appositions with no spine , as illustrated in Fig . 1b . Note that we do not claim that the functional distinction between thin and large spines is actually as clear-cut as assumed in the model – the model merely represents a coarse-grained spine state . In this model , transitions between the three morphological/functional states are possible . Following [22] and [50] , such transitions can occur either due to intrinsic fluctuations , or depending on the activity of the pre- and postsynaptic neuron . Note also that , although the inactive synaptic contact is allowed as a transitional state , it turns out to rarely occur in the optimized models that will be discussed below . The basic idea of the model put forward in this study is the following: As described in [27] dendritic spines have the biomolecular capability of detecting correlations in the relative spike timing of the pre- and postsynaptic neuron . If there are several active excitatory synaptic contacts from a presynaptic neuron to a single postsynaptic cell , all these synaptic contacts contribute to elicit spikes in the postsynaptic neuron . Hence each of the contacts increases the correlation between the two cells , measurable at each of the corresponding dendritic spines . So even if there is no direct communication between the synaptic contacts , they affect each other indirectly by increasing the correlation of pre- and postsynaptic spikes . Spike-timing dependence of structural plasticity is thus a candidate mechanism for the cooperation between synaptic contacts . According to [27] the magnitude of calcium influx into the dendritic spine depends on the proximity of pre- and postsynaptic spikes in time . The calcium influx activates or deactivates CaMKII macro-molecules and thus leaves a local memory . We call such a memory of the spike-timing correlation a correlation trace . The activation of the CaMKII subunits can be preserved for a long time [51] . The model we consider here , however , does not rely on the biophysical details of CaMKII activation , but just assumes a correlation trace is available . For the purpose of this study , the correlation trace could also come about by other mechanisms . Employing this correlation trace , we introduce a phenomenological model for activity dependent maturation , shrinkage and pruning of spines depending on the correlation of the spike-timing of pre- and postsynaptic cell , as described in detail in Methods . The model is based on [22] and incorporates the basic properties of structural plasticity [21] , activity-independent creation and pruning of spines , intrinsic fluctuations of spine volume , and activity-dependent spine remodeling . The set of all synaptic contacts connecting a given pair of neurons constitute a synapse , see also Fig . 1a . The state of a synapse is defined by the number of active contacts ( large spines ) and inactive contacts ( thin spines ) . The time-evolution of the synapse state is then described as a Markov process . For a given parameter set , we solve for the stationary probability distribution of the states . The parameters of the model were then optimized so that the distribution of the total number of synaptic contacts reproduces the experimental reference data , shown in Fig . 2a1–c1 , along with the respective transition rates of the model ( 6 ) in a2–c2 . For each of the three reference datasets ( a , b , c ) , we show the best model that resulted from the optimization . The models can reproduce the experimental distributions of synapse numbers . The existence of such a stationary distribution means that the average numbers of inactive , active and potential sites of the synapse do not change in time . This is so despite the constant creation and pruning of synaptic contacts since these processes compensate each other in equilibrium . Implicitly , the model allows that inactive , active and potential sites coexist between a given pair of neurons . The parameter sets for the displayed models are given in the second section of Tab . 1 . The fit of the connection L4–L23 takes very different parameter values than the others . Nonetheless for all three modeled connections , the time constant of the correlation trace is large compared to the time scale of fluctuations of neuronal activity , in agreement with our assumption about the distribution of the correlation trace . Concerning the parameters of the activity dependent structural plasticity , we find qualitatively similar results across datasets: In all three cases , both maturation and shrinkage/pruning rates decrease with increasing active synapse number , granting long-term stability to established synapses . The models of the intralaminar connections L4-L4 and L5-L5 show remarkable similarities . Both have a comparable and the rate of intrinsic , activity independent transitions is low , although this was not an a priori assumption . The parameter values for and are difficult to interpret individually . Across all the models , inactive synaptic contacts are rare , as indicated by the fraction of which ranges between and and . A fit of both connections L4-L4 and L5-L5 with a single parameter set is displayed in Fig . 3 . Although the model distributions in Fig . 3a1 , b1 do not follow the reference data as closely as in Fig . 2 , a good agreement of the distributions and the reference is achieved . For each of the three reference datasets we obtained many models with a comparable fit error . Fig . 2a4–c4 and Fig . 3a4 , b4 show the error ( circles ) of the best parameter sets obtained , ordered by the value of . We also investigate how the model distribution changes in response to an increase in the baseline probability of causal spike pairing . Some models decrease their contact number , while other models increase it , as can be seen from the derivative ( squares in Fig . 3 , 1 , row 4 ) . This quantity can take very different values for comparable fit errors . We call a model Hebbian if the number of active contacts grows upon an increase of causal spike pairings ( ) . Conversely we call a model anti-Hebbian if the number of contacts decreases ( ) . This diversity indicates that the plasticity model used here is general enough to implement Hebbian and anti-Hebbian learning , depending on the parameters . In Tab . 1 the value of the derivative is given along with other properties of the selected models . For each dataset , we selected the best model irrespective of it being Hebbian or anti-Hebbian . Another characteristic property of a model is the joint distribution of active synapses and inactive synapses ( Fig . 2 and 2 , row 3 ) . Especially in the model connections L4-L4 and L5-L5 and tend to be strongly correlated . While the expectation value of is largely determined by the reference data , the smaller expectation value of indicates that only a small proportion of spines are small and functionally weak . The marginal distribution of active and inactive synapses are shown for the three best models in Fig . 4a . In all selected models , across the datasets , the expectation values of and do not sum up to the expectation value of the number of contacts , which means that many unrealized synapses ( close appositions without inactive or active contacts ) are present , consistent with experimental findings [21] . Fig . 4b addresses the question whether a homeostasis of the neuronal firing rate can be achieved by this structural plasticity model . Here , a homeostasis means that an increase in firing rate leads to pruning of input synapses , thus lowering the firing rate in effect . Conversely a decrease in firing rate should lead to synapse maturation . If that is the case , the plasticity rule establishes a homeostatic control of the firing rate to a fixed value . In our model , a negative derivative means that the plasticity rule acts as a firing rate homeostasis . From Eq . ( 5 ) we conclude that a change in can either increase or decrease the value of for a given , depending on the value of the baseline correlation . Previously , we arbitrarily set to . However , given a parameter set , we can change the value of to without changing the transition rates ( and all equilibrium properties of the model ) if we also shift the thresholds to ( 27 ) because then all the distances to the threshold are preserved . Thus is effectively a free parameter of the model and can be adjusted to set , as is shown in Fig . 4b . Hence the structural plasticity model we propose can establish a firing rate homeostasis . Furthermore we derived the expected lifetime of a synapse in the model , which is also shown in Tab . 1 . Here the lifetime is defined as the expected time until the synapse is pruned . Before being pruned , it can go back and forth between the states inactive and active several times ( fluctuate in volume ) . The lifetime is very different for inactive and for active synapses , the latter exceeding the former by about one order of magnitude or more . This is due to the fact that typically several active contacts coexist and mutually stabilize , which entails small rates , and ( cf . Fig . 2 row 2 ) . If an active synapse becomes inactive , the rate to go back to the active state also increases , which promotes going back and forth through these states . This behavior matches nicely with the volume fluctuations of large dendritic spines described in [22] . Through ( 25 ) finally we can relate the time scale of the models to experimental data [10] . The values for are listed in Tab . 1and confirm our assumption of a time scale separation of structural plasticity and neuronal activity . Using this estimate of the timescale , the lifetimes of inactive contacts are about a couple of days , while the lifetimes of active contacts span from a month up to years . [10] called spines with a lifetime of less than days transient , and spines with longer lifetimes persistent . This distinction roughly applies to the lifetimes of inactive and active contacts in our model .
We propose a model of structural plasticity to explain cooperative synapse formation [20] . The transitions of the states of synapses are assumed to depend on a signal locally available to a spine that depends on the correlation between pre- and postsynaptic activity , the correlation trace . There is strong evidence that a correlation trace could indeed be implemented in the dendritic spine through phosphorylation of the macromolecule CaMKII [35] , [27] , [34] . CaMKII has also been shown to be necessary for structural and long-term plasticity [52] , [53] , [4] , and may also drive presynaptic changes [54] . Here we assume an abstract , effective correlation trace instead of explicitly modeling the dynamics of CaMKII . This makes our results independent of the specific mechanisms employed at the synaptic contact , since also other processes may be available to form the correlation trace . We assume the correlation trace at the spine is forgetful , such that it integrates causal and anti-causal spike pairing events like a leaky integrator with a certain time constant . This time constant affects the location of the equilibrium probability distribution of the correlation trace and its variance . Across the datasets L4-L4 and L5-L5 , the time constants are comparable . If the correlation trace is implemented biologically by the cycle of expression , activation and degradation of CaMKII , these time constants will be observable in experiments . The optimized values for the time constant are well in the range of possible values that sustained CaMKII activation can show [51] for all three reference datasets . The differences in the model parameters of the connection L4–L23 compared to the other two intralaminar connections might be explained by the finding that most synaptic contacts of this connection are formed on dendritic shafts rather than on spines [55] . At dendritic shafts functionally similar plasiticity mechanisms could be at work , but our model might be less appropriate for this type of connection . However , although in early postnatal development more shaft synapses exist , in later stages synapses on spines dominate [56] , [57] . The rates of structural changes at the synapse are assumed to be a function of the equilibrium correlation trace distribution . To model this dependence mathematically we chose a versatile functional form ( 7 ) . This is necessary since a comprehensive quantitative description of the correlation dependence of structural plasticity is not known to date . Our optimization results for the transition rates show a strong selectivity for specific numbers of active contacts in a synapse: Transition rates are much higher in case there are few active contacts between two neurons , and many active contacts stabilize the system in all of the three modeled intra-cortical synapse types . Future experiments could investigate whether synaptic contact number ( or EPSP amplitude ) correlates with calcium transient amplitudes at the spines and with rates of spine maturation , shrinkage and pruning . Using the optimized models we also computed the expected lifetimes synaptic contacts . The lifetime of active contacts is about ten to one hundred times larger than the lifetime of inactive contacts across our models . This can be understood given the experimental references' results that an active contact is always accompanied by several others . For such synapses , our models predict a vanishing rate of activity dependent transitions , which lets the synapses stay in the active state for a long time . Thus persistent spines here correspond to active contacts , and transient spines to inactive contacts . Our finding constitutes a statistical explanation of the existence of these two distinct classes of spines [10] . Our best-fit models show functional differences . Most notably , the models can be either Hebbian or anti-Hebbian , in the sense that an increase in the frequency of causal spike pairing leads to either increased or decreased numbers of active contacts . Both Hebbian and anti-Hebbian connections have been observed in the neocortex [58] . For all connections we found comparably good fits of both types . Furthermore our model predicts a joint probability distribution of active and inactive contacts which goes beyond current experimental references . Future experiments which determine both of these numbers for many neuron pairs will allow further evaluation of our model . A possibility to optically distinguish and monitor active and inactive synapses in experiments might be to use fluorescent markers for AMPA and NMDA receptors . Synaptic contacts that what we call “inactive” should show less AMPA than “active” ones , but the inactive ones also include those synapses with few AMPA receptors . Previous models of structural plasticity have assumed a homeostasis of the firing rate [59] , [3] , in the sense that if neuronal activity increases beyond an a-priori chosen set-point , synaptic contacts are pruned to decrease the excitatory drive , and the reverse for activity below the set-point . Indeed the correlation dependent structural plasticity model [27] shows this behavior . We have investigated whether our models show firing rate homeostasis by computing how the expected number of active contacts changes with the firing rate . This dependency can be chosen arbitrarily by adjusting a free parameter of the model ( see Fig . 4b ) . Our model hence is capable of providing the proposed firing-rate homeostasis for properly chosen parameters . To obtain a simple Markov process , we used the discrete categories “unrealized” , “inactive” and “active” to describe the state of a synaptic contact . Technically our model is similar to the cascade synapse model of [60] but adds the morphological interpretation of the synaptic states . The inactive contact might be closely related to silent synapses , but in the actual biological system such a clear-cut distinction between functional states can probably not be made , see for example [38] . Busetto et al . [61] found that silent synapses are abundant in the developing animal but vanish in the adult . However , only spines that were morphologically mature were included in their study , making no claim about existence of thin spines with small heads . Quantal EPSC analysis in the adult neocortex showed that close to all synaptic contacts of the connection L4–L23 are functional [55] . Our model of this connection also shows no inactive synapses in expectation , which renders them unobservable in practice . Further [62] find in cultured hippocampal slices that newly formed spines contain AMPA receptors . Small spines , however , are generally easy to miss , since they are often smaller than the resolution limit of optical microscopy [10] , [61] , and they may also be pruned again quickly after formation [63] . After all there is ample evidence that newly formed spines are small [21] and that AMPA receptor density correlates with volume [22] . We thus follow [43] and approximate thin , small spines as inactive synaptic contacts , and large spines as active ones as described above in detail . As a consequence of the coarse-grained description of the state of synaptic contacts , all active synaptic contacts in our model produce an EPSP of a fixed amplitude . However , in biology this amplitude varies from contact to contact . Including a fine grained description of synaptic amplitudes in a structurally similar model as the one presented here would result in a massive increase of the dimension of the state space and is therefore potentially unfeasible . Such a dispersion of synaptic amplitudes would result in a different functional dependence of the mean ( 2 ) and variance ( 3 ) of the correlation trace on the number of active contacts . However , at a given synapse the mean would still be monotonically increasing with . On a population level , the dispersion of synaptic amplitudes thus results in an additional contribution to the width of the distribution of the correlation trace in ( 3 ) . We can think of part of the noise added to as representing this contribution . This reduces the precision of correlation detection at the dendritic spine . In a model with dispersion of synaptic amplitudes , we therefore expect to find qualitatively similar fits for our coarse grained model at a correspondingly reduced additional noise . We defined that inactive synaptic contacts host NMDA receptors . The conductance of NMDA receptors increases upon a postsynaptic depolarization if the magnesium block is removed . At negative voltages NMDA channels have a smaller but non-vanishing conductance and hence mediate excitatory postsynaptic currents ( EPSC ) . However , the time scale of NMDA activation is much slower than that of AMPA channels . A postsynaptic action potential partially caused by NMDA currents of one synaptic contact would thus occur much later than the presynaptic glutamate release . The postsynaptic depolarization is therefore less efficient in opening the NMDA receptors at another synaptic contact of the same synapse . This , however , is the crucial mechanism that allows correlation detection and cooperation in our model . Hence one may assume that NMDA currents contribute much less to the correlation trace , and thus have vanishing impact on the cooperative plasticity of our model . We therefore use the term “inactive” here in a functional sense . In neonatal rat hippocampus also presynaptically silent synapses have been observed , which show a very low probability of transmitter release [64] , [65] . However , even a low probability of release enables the formation of a postsynaptic correlation trace at the dendritic spine as in our model . Moreover , even presynaptic changes of the transmitter release have been reported to depend on such a correlation trace in a similar way [54] . The dependence of maturation and shrinkage/pruning on the correlation trace that we use here is a sufficiently generic model to also include these presynaptic mechanisms , although we do not intend to model them here explicitly . The term structural plasticity describes a broad range of phenomena , many of which have not been addressed here . Competition between synapses from distinct neurons to a common postsynaptic neuron has been shown to be important for the emergence of cortical network structure [66] . In the more detailed models of structural plasticity in neuronal networks based on the activity of CaMKII [27] , [67] , cooperation and competition between synaptic contacts necessarily occurs . Here we assumed that synapses between different pairs of neurons develop independently , so inter-synaptic competition effects were not considered . Furthermore , structural plasticity also includes changes to the network structure that can come about by migration of axons on much longer time scales . Our model rather describes the steady state of the adult cortex , during which spines form and retract , but the axonal arborization can be assumed to be constant [13] . In lesion studies it has been shown that the steady state can become unstable and axons again begin to migrate [68] . Although simple and abstract in its description of complex cellular phenomena , our model can explain the cooperation of synaptic contacts in the adult neocortex , postulated in [20] . The model shows how continuously active structural plasticity can lead to the global configuration of synaptic contact numbers that was observed experimentally . The key ingredient of the model which mediates the necessary cooperation is a trace of the spike-timing correlations of the pre- and postsynaptic neuron . The resulting synaptic learning rule is local ( it solely requires mechanisms at the synaptic contacts ) but can nonetheless explain cooperative synapse formation . | Structural plasticity has been observed even in the adult mammalian neocortex – in seemingly static neuronal circuits structural remodeling is continuously at work . Still , it has been shown that the connection patterns between pairs of neurons are not random . In contrast , there is evidence that the synaptic contacts between a pair of neurons cooperate: several experimental studies report either zero or about 3–6 synapses between neuron pairs . The mechanism by which the synapses cooperate , however , has not yet been identified . Here we propose a model for structural plasticity that relies on local processes at the dendritic spine . We combine and extend the previous models and determine the equilibrium probability distribution of synaptic contact numbers of the model . By optimizing the parameters numerically for each of three reference datasets , we obtain equilibrium contact number distributions that fit the references very well . We conclude that the local dendritic mechanisms that we assume suffice to explain the cooperative synapse formation in the neocortex . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [
"developmental",
"neuroscience",
"synaptic",
"plasticity",
"neuroanatomy",
"neural",
"networks",
"computational",
"neuroscience",
"biology",
"neuroscience",
"learning",
"and",
"memory"
] | 2012 | Spike-Timing Dependence of Structural Plasticity Explains Cooperative Synapse Formation in the Neocortex |
This RDT might be used as a point of care diagnostic tool in limited resources countries . An evaluation in field conditions and in other epidemiological contexts should be considered to assess its validity over a wider range of serogroups or when facing different endemic pathogens . It might prove useful in endemic contexts or outbreak situations .
Leptospirosis is a bacterial disease of high incidence in many tropical and sub-tropical areas [1] , [2] . Its clinical presentation is both highly variable and most often characterized by non-specific signs and symptoms; the complete triad of Weil's disease ( hepatic failure , renal failure and hemorrhage ) is recognized to account for less than one third of human cases [3] , [4] . Most of the early signs and symptoms point to the so-called “acute fever of unknown origin” ( FUO ) , a major diagnostic challenge in tropical and subtropical areas . Beside the epidemiological context and patient exposure history , to quickly diagnose and implement an appropriate treatment , an etiological investigation is necessary , especially in malaria , hantavirus , or viral hepatitis endemic regions or during influenza , chikungunya or dengue outbreaks . Leptospirosis is also reported to be an emerging or re-emerging disease in industrialized countries , with probable increasing impacts due to global warming and increasing travel-related cases [5] , [6] . Unfortunately , the biological confirmation of leptospirosis is both tedious and rarely available in a timely manner . It notably requires sophisticated techniques that are most frequently available only in central reference laboratories [1] . These techniques are of prime importance in disease surveillance and epidemiological investigations but are inappropriate for early clinical care in peripheral health centers that support a major part of the leptospirosis burden . Additionally , an early and proper antibiotic treatment is known to be a key to rapid recovery and a major determinant of outcome in leptospirosis [3] , [7] , [8] . The need for portable rapid diagnostic tests ( RDT ) is striking and largely recognized to improve clinical management of leptospirosis patients , notably in remote dispensaries of tropical and subtropical regions . As an example , the major part of the leptospirosis burden in New Caledonia occurs away from the city and the central hospital of Noumea [9] , [10] . In this study , we developed a vertical flow immunochromatographic RDT for the early diagnosis of leptospirosis to detect Anti-Leptospira human IgM and evaluated its sensitivity , specificity , reproducibility , repeatability , temperature stability and simulated shelf-life in the context of leptospirosis endemic ( New-Caledonia and French West Indies ) or non-endemic ( mainland France ) countries using clearly defined case definitions and the reference Microscopic Agglutination Test ( MAT ) results as the gold standard [4] .
The antigen was prepared at the Institut Pasteur , National Reference Center for Leptospirosis ( Paris , France ) as follow: a culture of L . fainei serovar Hurstbridge strain BUT 6T at OD = 0 . 5 at 420 nm was killed by 0 . 2% formalin for 3 hours at ambient temperature , boiled for 45 minutes and its pH adjusted to 9 . 6 . This preparation was kept at 4°C until use as an antigen for the Vertical Flow RDT and was also used for an in-house IgM ELISA at the National Reference Center [11] ( Bourhy et al . , manuscript in revision ) . The positive control line was made of purified human IgM ( MP Biomedicals ) at 2 mg/mL . Both control and test lines were sprayed as lines onto nitrocellulose membranes . Gold particles labelled with goat anti-human IgM ( BBI International BA . GAHM40/X ) adjusted to the concentration of OD520 nm = 3 were used as the capture mobile phase to construct our one-step vertical flow immunochromatography RDT , as previously described [12] . Two batches of RDT were produced at the Institut Pasteur in Paris ( platform 5 ) and used in this study: batch numbers 110 , 211 and 120 , 511 . Leptospirosis cases were defined as confirmed when a clinical and epidemiological suspicion was complemented by either a positive specific PCR evidencing genomes of pathogenic Leptospira spp . in the blood or urine of the patient , or when the MAT on acute and convalescent sera showed a seroconversion ( from nil to a titer ≥400 ) or a significant seroascencion ( at least a fourfold raise in titers ) [13] . Because most of the serum specimens originated from leptospirosis endemic regions where a high MAT titer threshold is usually used , we adopted this ≥400 threshold throughout the study . Probable cases were defined as clinical and epidemiological suspicion together with a unique serum with a MAT titer ≥400 . The panels of strains used for MAT were adapted to the local epidemiology of New Caledonia , mainland France and French West Indies [9] , [14] and are provided in details in Table S1 . High rates of agglutination of the serum with one particular strain are used to identify the presumptive serogroup of the infecting bacterium as described elsewhere [4] . All sera used in this study were addressed to the Institut Pasteur in Nouméa or Paris for diagnostic purpose and originated from patients from New Caledonia , mainland France or the French West Indies ( Martinique and Guadeloupe ) . The workflow is summarized as a flowchart in Figure S1 . These laboratories are territorial ( New Caledonia ) and national ( France and West Indies ) reference centers for leptospirosis and receive all ( New Caledonia ) or more than 75% ( France and French West Indies ) of leptospirosis diagnosis requests . Sera were stored at −20°C , selected according to case definitions , and then tested blindly with RDTs . To assess the sensitivity of the RDT , only MAT-positive sera from confirmed cases were used: we tested 187 confirmed leptospirosis cases sera with a MAT titer ≥400 ( 120 from New-Caledonia , 38 from mainland France , 29 from French West Indies ) . The specificity was assessed using 221 sera ( 142 from New-Caledonia , 79 from mainland France ) : 12 anti-Chikungunya virus IgM positive sera , 58 anti-dengue virus IgM positive sera from all 4 serotypes , 6 anti-hepatitis A virus total Ig positive sera , 7 rheumatoid factor positive sera , 25 syphilis ( TPHA and VDRL ) positive sera , one acute malaria serum and 112 sera from healthy blood donors ( Platform ICAReB , Institut Pasteur ) . These 221 sera were then confirmed to be MAT-negative ( titer<100 ) within the same week . Preliminary experiments determined dilution of sera between 1/100 and 1/800 in Phosphate Buffer Saline ( PBS , pH 7 . 4 ) as suitable for testing with RDT . Because rapid tests are mostly to be used in endemic regions and similarly to the high MAT titer threshold , a 1/400 dilution of sera was used throughout the study . Briefly , Vertical Flow RDT strips were introduced into 200 µL diluted serum in 5 mL polystyrene tubes , for 15 minutes . The strips were then removed and placed on absorbent paper and read within 5 minutes . All results were recorded using a grading scale from 0 ( no visible trace on test band ) to 3+ ( intensity of the test band equal to the intensity of the control band ) . The grading included a “weak” value for low but visible traces on the test band . Weak , 1+ , 2+ and 3+ were then considered positive for further analysis and 0 was considered as negative . All analyses were run blindly: any person involved in one particular analysis had no access to the results of the other tests results from the same serum . To assess the sensitivity of the RDT , only the first MAT-positive serum from each confirmed case was used . For specificity evaluation , all 221 negative control sera have been tested using MAT and were all negative ( titer<100 ) ( see Table 1 ) . Possible false negative results due to high levels of anti-Leptospira IgM ( prozone phenomenon ) was controlled using two positive sera with 25 , 600 and 51 , 200 MAT titers: serial two-fold dilutions of the sera ( 1/400 to 1/6 , 400 ) were tested and test band intensity of the Vertical Flow RDT were compared . The strips were stored at 4°C in sealed aluminium foils with a silica gel bag to avoid exposition to humid conditions , and the test was performed at laboratory room temperature ( 20°–25°C in New Caledonia and Paris ) . The predictive shelf life of the RDT was assessed by testing serial dilutions of a MAT-positive serum ( MAT titer = 800 , pointing to serogroup Icterohaemorrhagiae ) twice per week over a period of 3 weeks exposure of the strips at 60°C . During this period , the positive control serum was kept at 4°C to avoid repeated freeze-thaw cycles . This accelerated stability method is equivalent to two years of actual storage time at 25°C [15] , [16] . Several experiments were performed in New Caledonia to evaluate the the Vertical Flow RDT . For reproducibility and repeatability , four sera ( 3 positives and one negative ) were tested blindly by three different operators on three different days; one serum was tested blindly 14 times by integrating 14 aliquots of the same specimen within 4 independent experiments and using two different batches of RDTs by the same operator; 177 sera from both confirmed and probable cases ( 28 negative and 149 positive samples ) were read independently by two technicians , including 157 sera ( 16 negative and 141 positive samples ) by three technicians . We additionally simulated tropical field conditions by performing the tests in parallel at 37°C in an incubator ( simulating tropical conditions ) and at laboratory temperature ( standard condition ) The earliness of IgM seroconversion using MAT or RDT was assessed on serial sera ( day 2 to day 18 after the onset of symptoms ) from 17 confirmed cases , based on the date of onset as declared by the patients . The RDT was also tested on early sera from 99 patients who were tested positive by PCR but negative by MAT ( titer = 0 , 100 or 200 ) . One hundred and fifty MAT-positive sera from probable cases , including 124 sera from the IPNC collection and 26 from the French National Reference Centre , were tested using RDT . To compare the newly developed RDT with currently available techniques , we compared its performance on identical sera from New Caledonia . To assess the sensitivity , 72 MAT-positive sera from confirmed cases were randomly selected from the 118 New Caledonian control sera . For the specificity , 72 negative controls were randomly-selected , corresponding to 10 anti-Chikungunya virus IgM positive sera , 30 healthy blood donors , 11 anti-dengue virus IgM positive sera from all 4 serotypes , 6 anti-hepatitis A virus total Ig positive sera , 7 rheumatoid factor positive sera , 7 syphilis ( TPHA and VDRL ) positive sera , one acute malaria serum . The results using our RDT were compared with those obtained using two Elisa assays ( Leptospira IgM ELISA , Panbio , Inverness Medical , QLD Australia , and SERION ELISA classic Leptospira IgG/IgM , Institut Virion/Serion GmbH , Germany ) and one lateral flow IgM immunochromatography assay ( Leptocheck , Zephyr Biomedicals , India ) . The Serion ELISA test was used together with the Rheumatoid Factor Absorbent as recommended by the manufacturer . All tests were made within a 5 day period . For calculations , the “uncertain” results of ELISA were considered as positive . The evaluation of our RDT for the serodiagnosis of leptospirosis was performed according to the WHO Tropical Diseases Research Diagnostics Evaluation Expert Panel for the evaluation of diagnostic tests for infectious diseases [17] . Data were captured into Excel 2007 ( Microsoft Corporation , Redmond , United States of America ) . We calculated sensitivity ( Se ) , specificity ( Sp ) , positive and negative predictive values ( PPV and NPV , respectively ) of the RDT , using the reference MAT serology as the gold standard . The 95% confidence intervals ( 95% CI ) were calculated using the Wilson's method . The variations of the PPV and NPV according to the prevalence of the disease were also plotted . We also calculated likelihood ratios ( LR ) . The positive LR ( LR+ = Se/[1 - Sp] ) indicates how many times a positive result is more likely to be observed in specimens with the target disorder than in those without the target disorder . The negative LR ( LR− = [1 - Se]/Sp ) indicates how many times a negative result is more likely to be observed in specimens with the target disorder than in those without the target disorder . The more accurate the test is , the more LR differs from 1 . LR+ above 10 and LR− below 0 . 1 are considered convincing diagnostic evidence [18] . The 95% CIs were calculated for LR+ and LR− [19] . The diagnostic odds ratio ( DOR ) measures the test performance by combining the strengths of sensitivity and specificity , with the advantage of representing a single indicator of accuracy . These characteristics make the DOR particularly useful for comparing tests whenever the balance between false negative and false positive rates is not of immediate importance [20] . The DOR is defined as the ratio of the odds of positive test results in specimens with the target disorder relative to the odds of positive test results in specimens without the target disorder . It was calculated as follows:The DOR does not depend on prevalence and its value ranges from 0 to infinity , with higher values indicating better discriminatory test performance . The 95% CIs for DOR values were also calculated [21] . The IPNC and the French NRC are reference diagnostic laboratories for leptospirosis . In New Caledonia , leptospirosis is a notifiable disease . The serum samples used in this study were selected from the 2008–2001 IPNC and 2009–2011 French NRC collections of sera issued from routine diagnostic activities and as part of public health surveillance . This biobank of sera was declared to the French Ministry of Research ( DC-2010-1222 , Collections number 1 and 2 ) . This study was part of a protocol approved by the Institut Pasteur ( protocol # RBM2008-16 ) and the French Ministry for Education & Research ( protocol # AC-2007-44 ) . All sera were tested as anonymous samples . Negative sera from mainland France were provided by Platform ICAReB ( Investigation Clinique et Acces aux Ressources Biologiques ) . The STARDT checklist is provided as Table S2 .
The sera used in this study were from patients or donors of both sex and all age classes , being selected among leptospirosis suspicions ( positive sera ) , other pathologies or blood donors ( negative sera ) . All along the study and whatever the batch used , we observed no invalid test: all RDT displayed an intensely marked control line and very little to no background coloration . Out of the 187 gold standard positive sera tested , 168 were RDT positive , including 15 RDT with a test line intensity graded as “weak” ( 8 . 9% of positives ) . The putative serogroups of the 19 RDT negative sera were: Icterohaemorrhagiae ( n = 12 ) , Pyrogenes ( n = 3 ) , Australis ( n = 2 ) , Panama ( n = 1 ) and one could not be determined due to co-agglutination of multiple serogroups . Out of the 221 MAT-negative sera tested , 207 were RDT negative . All 14 RDT positive sera were graded “weak” and originated from 9 healthy blood donors and five patients positive for anti-dengue virus IgM . The sensitivity and specificity of the RDT were therefore , respectively , Se = 89 . 8% [95% CI , 84 . 7–93 . 4] and Sp = 93 . 7% [95% CI , 89 . 65–96 . 2] . The Likelihood Ratios ( LR ) were therefore LR+ = 14 . 18 [95% CI , 8 . 52–23 . 56] and LR− = 0 . 11 [95%; 0 . 01–0 . 17]; and the Diagnostic Odds Ratio DOR of 130 . 74 [95% CI , 63 , 65–268 , 52] . The results are summarized in Table 1 , positive and negative predictive values of our RDT according to prevalence are presented in Figure 1 . The absence of false negative due to prozone phenomenon was demonstrated using dilutions of sera with highest MAT titers: serial two-fold dilutions actually yielded test lines of decreasing intensities . . To simulate tropical conditions , the RDT results of 10 MAT-positive sera run at 37°C were compared and proved identical to those run at 25°C . For accelerated shelf-life evaluation , serial two-fold dilutions ( from 1/400 to 1/12 , 800 ) of one MAT positive serum ( titer 800 ) were tested twice a week for three weeks with RDT stored at 60°C . At day 1 , the RDT was positive at a 6 , 400 dilution , and remained the same until day 17 . At day 21 , the 3 , 200 dilution was the last giving a positive test line ( a one dilution decrease of the serum ) . One serum tested 14 times with strips from the two different batches gave 14 similar results , including the grade of the test line . Inter-readers variability was assessed by two independent operators on 177 sera ( 28 negative and 149 positive ) of which 157 sera were read by three independent operators . These readings provided an excellent inter-operator agreement ( >99% ) in all cases but one: one weakly positive RDT from a probable case was rated “weak” by two operators but negative by the third one . Inter-operator variability was also assessed using 4 sera ( RDT graded from negative to 3+ ) blindly and independently tested on three different days by three different operators . Two operators provided perfectly concordant grading results on all three tests , the third one graded “weak” a negative serum once out of the three tests . Of 17 confirmed cases analysed ( see Table 2 ) , one patient ( number 1 ) seroconverted for MAT at day 6 ( pointing to Icterohaemorrhagiae ) but remained negative for RDT; oppositely , 5 PCR-confirmed patients ( numbers 2–6 ) were MAT negative whereas they were RDT-positive . For one of these patients ( number 6 ) , PCR and RDT tests were both positive at day 4 after onset of symptoms . Five patients ( numbers 7–11 ) were positive for MAT and RDT on the same day ( days 5–11 after the onset of symptoms ) ; lastly , for 6 patients ( numbers 12–17 ) , the RDT was positive earlier than the MAT ( day 3 to day 7 ) . Out of these 6 , four ( numbers 12 , 15 , 16 and 17 ) had a positive blood PCR and RDT results on the same day ( on days 5 , 4 , 7 and 3 respectively ) . Similarly , in 16 out of 99 early sera from confirmed patients from New Caledonia , the RDT was positive whereas the MAT was still negative ( 6 out of 62 MAT negative ) or displayed low titers ( 4 out of 21 with a MAT titer of 100; and 6 out of 16 with a MAT titer of 200 ) . Of 150 sera from probable cases of leptospirosis ( unique sera with a MAT≥400 ) , 109 gave a positive result using the RDT , corresponding to a concordance of 72 . 7% [65–79 . 1] . Out of these , 108 had a MAT>400 , from which 81 ( 75% [66 . 1–82 . 2] ) were RDT-positive , while 63 had a MAT titer >800 , from which 53 ( 84 . 1% [73 . 2–91 . 1] ) were RDT- positive . The use of 72 gold standard positive ( MAT≥400 from confirmed cases ) and 72 negative ( MAT<100 ) serum specimens selected randomly allowed a comparison of our RDT with three commercially available tests: two ELISA tests ( from Serion ( using the recommended RF absorbant ) and from Panbio ) and one IgM lateral flow immunochromatographic assay ( Leptocheck ) . The results of these tests are detailed in Table 3 . The IgM ELISA from Panbio had 100% specificity on these particular specimens together with the lowest sensitivity ( 75% ) . This 100% specificity does not allow the calculation of a Diagnostic Odds Ratio ( DOR ) that would however be very high . The ELISA test from Serion had both a good sensitivity ( 91 . 7% ) and a good specificity ( 81 . 9% ) , therefore showing a good DOR of 49 . 9 . Another rapid diagnostic test , namely Leptocheck ( from Zephyr ) had a very good sensitivity ( 91 . 2% ) but a quite low specificity ( 52 . 8% ) , giving a DOR of 39 . 1 . The Vertical Flow RDT we developed displayed a very good specificity ( 95 . 8% ) and a good sensitivity ( 81 . 9% ) and had therefore a very good DOR of 104 . 4 . The corresponding curves of predictive values according to the prevalence of the two IgM rapid tests on these specimens are compared in the figure 2 .
Serum samples from leptospirosis patients contain antibodies that become detectable approximately one week after the onset of symptoms [4] . MAT is a long used gold-standard method for the serological diagnosis of leptospirosis . This method relies on the detection of agglutinating antibodies ( both IgM and IgG ) against antigens that are live Leptospira strains corresponding to representative serogroups of epidemiological significance . However , many laboratories and hospitals do not have the facilities and expertise required to perform the MAT . Therefore , there is a growing use of qPCR for early diagnosis [22] , providing the opportunity to rapidly confirm leptospirosis suspicions in acute phase sera . More simple and rapid serological diagnostic tests including ELISA-based assays detecting antibodies are also used . However , none of these techniques are prone to be implemented in health centers of endemic regions where the highest burden of leptospirosis occurs . Pathogenic Leptospira strains are classified into 9 species and more than 200 serovars reflecting the structural heterogeneity in the carbohydrate component of the lipopolysaccharide . ELISA-based assays using crude whole-cell lysates of Leptospira strain ( usually the saprophyte L . biflexa serovar Patoc strain Patoc 1 ) as antigens may not recognize the diversity of circulating strains and the sensitivity of these tests are generally poor [3] , [4] . A major challenge is to discover antigens that are conserved across the major leptospiral strains . In this study , we tested a one-step vertical flow immunochromatography RDT coated with heat-killed L . fainei serovar Hurstbridge as an antigen . L . fainei belongs to the intermediate group of Leptospira [23] and , as such , may share common antigenic features with saprophytes and pathogens which constitute the two other phylogenetic groups in the genus Leptospira . In addition , a previous study has suggested that L . fainei serovar Hurstbridge may cross-react with different pathogenic serovars [24] . Rapid diagnostic tests should ideally be accurate , simple to use , relatively inexpensive , easy to interpret , stable under extreme conditions , with little or no processing , and give the results within less than 2 hours . A proper evaluation of a diagnostic test has to face two major challenges: first , the samples used for validation must have a very well-defined status with regard to the diagnostic target of interest; second , the results of the test under evaluation must be compared with the results of the same samples characterized using a validated reference test defined as the “gold standard” . Our study evaluated the sensitivity and specificity of a new Vertical Flow RDT for the serological diagnosis of leptospirosis in endemic ( New Caledonia and French West Indies ) and non-endemic ( mainland France ) countries . We only used sera from confirmed leptospirosis cases [13] for this evaluation . Therefore , the positive samples for the evaluation of sensitivity were both gold standard positive ( a MAT titer of at least 400 ) and from confirmed leptospirosis cases ( either a positive PCR or a seroconversion from nil to ≥400 or a ≥4-fold rise in MAT titer in paired sera ) . This high MAT threshold was chosen because most of the specimens ( 149 out of 187 ) originated from endemic regions , where a similar MAT threshold is used for diagnosis and surveillance [4] , [9] . Additionally , negative sera for specificity evaluation were tested blindly using the reference MAT and were only considered as true negatives if the MAT titer was below 100 . These latter originated from both healthy volunteers and a selection of patients with pathologic conditions of relevance in endemic countries . Using this clearly defined case definition , sensitivity and specificity were assessed using a high dilution of sera ( 1/400 ) reflecting the high MAT threshold titer . The sensitivity and specificity of the new Vertical Flow RDT in these conditions were 89 . 8% and 93 . 7% respectively . These results compare and are slightly better than the ones reported by Smits et al . who reported a 85 . 8% sensitivity and a 93 . 6% specificity with another Vertical Flow RDT [25] . The performance of ELISA tests vary widely in terms of sensitivity and specificity . For example , a commercial IgM ELISA ( Panbio ) gave a sensitivity and specificity of 76% and 82% in northeast Thailand [26] , 35% and 98% in Hawai [27] , and 61% and 66% in Laos [28] , respectively . Reported variations in diagnostic assay performance may reflect population-related differences such as past exposure to leptospirosis or environmental leptospires . This can also be attributable to differences in the choice of the case definition; MAT is usually used as the reference test [29] . To increase the statistical power of our evaluation , we included as many serum samples as possible , only including the earliest positive serum when serial samples were available . Because the patient population recruited through our laboratories represents all ( New Caledonia ) or around 80% ( France and West Indies ) of leptospirosis suspicions , our collection can be regarded as representative of the total patient population in these regions . These included sera from New Caledonia older than 3 years , stored frozen at −20°C . It is well recognized that the long term storage of serum specimens at −20°C and their freeze/thawings may result in a drop of IgM titers . Actually , the sensitivity was higher in sera stored for less than two years than in sera stored for more than two years ( 90 . 6% versus 81 . 5% ) . This may have resulted in a slight under-estimation of the sensitivity of this Vertical Flow RDT . The RDT we developed reacts with IgM to at least serogroups Australis , Autumnalis , Ballum , Bataviae , Canicola , Cynopteri , Grippotyphosa , Hebdomadis Icterohaemorrhagiae , Panama , Pomona , Pyrogenes , Sejroe and Tarassovi , indicating that the assay reacts broadly with antibodies mounted against Leptospira strains circulating worldwide . Probable leptospirosis cases were defined as cases with a leptospirosis-compatible clinical presentation but a unique serology with a MAT titer ≥400 . We tested our Vertical Flow RDT using these unique sera from such cases . Interestingly , the proportion of Vertical Flow RDT -positive sera was significantly lower than the sensitivity of the test as determined using confirmed cases ( 69 . 4% versus 89 . 8% , χ2 = 9 . 08 , p<0 . 01 ) , suggesting a poorer positive predictive value of the RDT for patients classified as probable cases . Two main reasons might contribute in explaining this difference . First , our Vertical Flow RDT only detects IgMs whereas the MAT is known to detect both IgMs and IgGs . Because IgM titers are known to decline faster than IgG , some positive MAT results may reveal IgGs remaining from previous exposure to leptospires . MAT could therefore be less specific than IgM-specific assays to detect acute and recent leptospirosis . It is also well known that direct visual methods such as MAT ( agglutination of bacteria using microscopy ) are less sensitive than indirect amplified techniques such as ELISA or colloidal gold particles immunochromatography assays . The possibility of false positive MAT results was already observed in other contexts [30] . Actually , though MAT is the recognized reference technique for the serological diagnosis of leptospirosis , it also suffers some drawbacks and weaknesses . Some concerns about both its sensitivity and its specificity have been raised and discussed [30]–[36] . More recently , a mathematical modeling study again demonstrated the limitations of MAT as a gold standard [29] . Since no diagnostic assay is adequately sensitive and specific enough to diagnose all acute cases of leptospirosis , results should be confirmed by another method . Regarding sensitivity , the most widely recognized weakness of MAT is its low sensitivity in early acute phase sera . The ability to detect anti-Leptospira antibodies earlier in the course of the disease with specific IgM detection tests than with MAT has already been largely recognized [30] , [33] , [34] , [37]–[41] . We also observed an earlier positivity of our IgM Vertical Flow RDT when compared to MAT in serial sera . In our study , 11 out of 17 leptospirosis seroconversions could be diagnosed earlier with the Vertical Flow RDT than with the MAT ( Table 2 ) . Similarly , in 16 out of 99 early sera from confirmed cases , the Vertical Flow RDT was positive before the seroascension ( 10/37 ) or seroconversion ( 6/62 ) could be evidenced with the MAT . However , for strictness reasons , our strategy was to only use gold standard-positive specimens ( MAT titers ≥400 ) for sensitivity assessment . Therefore , the sensitivity evaluated here might not reflect the conditions of rapid tests use in routine medical conditions , where patients may be seen and tested before seroconversion . Previous studies have generally found that ELISA-based assays detect anti-Leptospira antibodies earlier in the course of the disease than with MAT [4] , [30] , [33] , [37] , [38] . Anti-leptospiral IgM cannot be detectable before 4–5 days after onset of symptoms , before the appearance of IgG and agglutinating antibodies [42] . Because an early diagnosis is of prime importance in the clinical management of leptospirosis [43] , the possibility to ascertain the disease earlier in the course of the infection should be regarded as a real asset . When considering the need of RDT for bedside diagnosis , the comparison of our Vertical Flow RDT with a RDT that is commercially available shows that our test has a lower sensitivity ( 81 . 9% versus 97 . 2% ) but a much higher specificity ( 95 . 8% versus 52 . 8% ) and therefore a better Diagnostic Odds Ratio ( 104 . 4 versus 39 . 1 ) . This better performance is also shown by the comparison of the curves of their predictive values according to prevalence ( Figure 2 ) . There may be various reasons for these differences , including a different antigen used for IgM detection . However , the most probable cause is a much lower dilution of serum specimens in the Leptocheck lateral flow IgM assay ( ca . 1/20 ) when compared with the dilution used for our Vertical Flow RDT ( 1/400 ) . Because the MAT positive threshold titers may vary depending on the region [44] , endemic countries usually using a higher threshold , it might be worth also considering using a different serum dilution for RDT according to the local epidemiology of leptospirosis . Because leptospirosis is endemic in New Caledonia and French West Indies , we decided to consider both a MAT titer of 400 as a gold standard and a 1/400 dilution of the serum for the Vertical Flow RDT evaluation . It is highly probable that the use of a lower dilution ( 1/200 or 1/100 ) would result in both an increased sensitivity and a decreased specificity . Our results demonstrate that this new rapid diagnostic test could prove useful in endemic contexts , especially in low and middle-income countries . Actually , most of the leptospirosis burden occurs in the back-country with delayed access to the reference laboratories . In epidemics situations , especially during post-disaster periods like in the Philippines in 2009 [45] , reference diagnostic tests are seldom if ever available . Therefore , a RDT with good diagnostic performances would also be particularly useful [1] . For easier use , further development of the technique could allow its use with capillary blood . However , because an early initiation of antibiotherapy is a major contributor to a rapid recovery , the recommendation of treating the patient on the sole basis of a clinical and epidemiological suspicion should be maintained . The use of this Vertical Flow RDT as an initial screen for leptospiral infections would still allow facilitating the difficult differential diagnosis of leptospirosis [46] . Lastly , because the MAT provides important epidemiological information at the population level , it should still be recommended that sera be sent to the reference laboratory for subsequent confirmation by MAT , as suggested by other authors [41] . | The major burden of leptospirosis happens in low-income populations from tropical or subtropical regions . Because of nonspecific symptoms in human leptospirosis , the biological confirmation is needed to ascertain the disease . However , this biological diagnosis relies on sophisticated and time-consuming techniques that are most often hardly ( if ever ) available to clinicians in peripheral health centers . Yet , the outcome of leptospirosis in humans largely depends on an early antibiotic treatment . Taken together , these factors highlight the need of rapid simple diagnostic tests for leptospirosis that could be used directly on the bedside even in remote health centers . In this study , we developed and evaluated a prototype point of care strip test for the serological diagnosis of human leptospirosis in New Caledonia , mainland France , and the French West Indies . The sensitivity was 89 . 8% [95% CI , 84 . 7–93 . 4] and the specificity 93 . 7% [95% CI , 89 . 65–96 . 2] . This easy , early and portable diagnostic test will be evaluated in other epidemiological conditions and under field conditions . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [
"medicine",
"infectious",
"diseases",
"test",
"evaluation",
"diagnostic",
"medicine",
"neglected",
"tropical",
"diseases",
"leptospirosis"
] | 2013 | Sensitivity and Specificity of a New Vertical Flow Rapid Diagnostic Test for the Serodiagnosis of Human Leptospirosis |
Metabolism is the major output of the circadian clock in many organisms . We developed a computational method to integrate both circadian gene expression and metabolic network . Applying this method to zebrafish circadian transcriptome , we have identified large clusters of metabolic genes containing mostly genes in purine and pyrimidine metabolism in the metabolic network showing similar circadian phases . Our metabolomics analysis found that the level of inosine 5'-monophosphate ( IMP ) , an intermediate metabolite in de novo purine synthesis , showed significant circadian oscillation in larval zebrafish . We focused on IMP dehydrogenase ( impdh ) , a rate-limiting enzyme in de novo purine synthesis , with three circadian oscillating gene homologs: impdh1a , impdh1b and impdh2 . Functional analysis revealed that impdh2 contributes to the daily rhythm of S phase in the cell cycle while impdh1a contributes to ocular development and pigment synthesis . The three zebrafish homologs of impdh are likely regulated by different circadian transcription factors . We propose that the circadian regulation of de novo purine synthesis that supplies crucial building blocks for DNA replication is an important mechanism conferring circadian rhythmicity on the cell cycle . Our method is widely applicable to study the impact of circadian transcriptome on metabolism in complex organisms .
The circadian clock represents a key time-keeping mechanism for many fundamental physiological processes such as cell cycle and metabolism . The cell cycle and the circadian clock are both well-studied periodic processes and their interaction has been a recent focus of circadian research . Until now , it has been observed that the circadian timing system gates cell cycle progression in various organisms from unicellular microorganisms such as Euglena gracilis [1–3] and cyanobacteria [4–7] to complex organisms such as humans [8] , mouse [9 , 10] as well as in in vitro culture systems [11] . These findings implicate the cell cycle as being an evolutionarily conserved regulatory target of the circadian clock . The disruption of the circadian timing system leads to cell cycle disorders and subsequently can result in various types of cancers [12–15] . Key circadian clock regulatory targets have been identified as important molecular links between the circadian system and the timing of cell division [9 , 16–19] . The progression of each phase in the cell cycle is mediated by the activity of cyclin-dependent kinases ( CDKs ) /Cyclins complexes and CDK inhibitors ( CKIs ) , which are directly or indirectly controlled by key circadian transcriptional factors [19–21] . Additionally , there are substantial lines of evidence suggesting that a number of key cell cycle regulators , such as p21 , p16 , cyclinE , cyclinA2 , wee1 , cyclinB1 , cdc2 are rhythmically expressed and clock controlled in many species [9 , 22 , 23] . Furthermore , recent studies showed that circadian and cell cycle oscillators were tightly coupled and synchronized in mouse fibroblasts in a 1:1 fashion at the single-cell level [24 , 25] . Zebrafish represents an excellent vertebrate model to study the interplay between the circadian clock and the cell cycle . Progression through S-phase and M phase is restricted to specific times of the light/dark ( LD ) cycle in a circadian-dependent manner in zebrafish larvae , adult tissues and even cell lines [22 , 26 , 27] . More recently the circadian clock has also been demonstrated to regulate cell proliferation and the expression of core cell cycle regulators in the zebrafish intestine [28] . The detailed mechanism of how the circadian clock orchestrates cell cycle has been studied intensively . Light can serve as an environmental signal that regulates cell division via the circadian clock system [27 , 29] . Glucocorticoids have also been implicated in playing a key role in the circadian control of cell cycle [30] . Furthermore , ADP , acting as a paracrine signal , has been associated with daily S phase cell activity in adult zebrafish retina [31] . Importantly , the direct clock control of mRNA expression of the cyclin-dependent kinase inhibitors p20 and p21 also clearly plays a critical role in cell cycle timing in zebrafish [32] . Recently , there has been considerable interest in how various metabolic processes are controlled by circadian clock [33] . In the meantime , a wealth of circadian transcriptome data has been generated by microarray or RNA-seq studies in different species . Therefore , it is imperative to develop methods to gain insight of the functional consequences of circadian transcriptome on metabolism in the context of global metabolic network . It is well-known that metabolism exerts a strong influence on cell cycle [34] . However , little is known about whether metabolic processes directly contribute to linking the cell cycle and circadian clock . Since the coordination of metabolic pathway activity is an essential prerequisite for major cellular events such as DNA replication during S phase [35 , 36] , we reason that circadian regulation of metabolism may represent a key control point for the cell cycle . In this study , we developed a new clustering method based on a distance measure incorporating both metabolic network and circadian phase information . We applied this method to the circadian transcriptome of zebrafish and systematically identified metabolic gene clusters showing coherent circadian gene expression . We found that the top gene clusters consist of mainly genes in purine and pyrimidine metabolism pathway . In this pathway , Inosine 5'-phosphate dehydrogenase ( Impdh ) catalyzes NAD+-dependent oxidation of inosine 5'-monophosphate ( IMP ) to xanthosine 5'-monophosphate ( XMP ) , which is the rate-limiting step in the de novo biosynthesis of GTP [37] . Interestingly , three homologous genes of impdh in zebrafish , namely impdh1a , impdh1b and impdh2 , all show robust circadian expression in both larval and adult zebrafish but reside in two different coherent circadian gene clusters . Consistent to this finding , we observed that the level of IMP also showed circadian oscillation in zebrafish larvae through metabolomics analysis . We showed that while impdh1a contributes to eye development and the synthesis and distribution of pigment , impdh2 plays an important role in the circadian control of cell cycle . Furthermore , impdh1b function serves to delay embryonic development , contrary to impdh2 function . The pharmacological inhibition of Impdh activity by mycophenolic acid ( MPA ) treatment attenuates the circadian control of cell cycle . Gene-specific knock down experiments revealed that the inhibition of impdh2 function is mainly responsible for this phenomenon . Thus , we hypothesize that purine synthesis is circadian clock-controlled via impdh activity and contributes to the daily rhythm of S phase in the cell cycle . Finally , in the light of findings from the mouse , we proposed that de novo purine synthesis may also mediate circadian rhythm of cell cycle in mammals . Taken together , our study suggests that circadian control of de novo purine synthesis represents a novel mechanism linking the circadian clock , metabolism and the cell cycle .
Our previous study has revealed that zebrafish start to show robust circadian activity in the larval stage at 5 days post-fertilization ( 5 dpf ) [38] . We have identified 2 , 847 zebrafish circadian oscillating genes ( ZCOGs ) under both LD and DD conditions in larval zebrafish data . In this study , we focus on the circadian regulation of metabolism . Due to a lack of a global metabolic network in zebrafish as in the case of human and mouse [39] , we compiled a zebrafish metabolic network from the information of metabolic reactions , enzymes , and genes of zebrafish from KEGG database [40 , 41] ( S1 Dataset ) . Such network provides global connectivity between enzymes or reactions through metabolites in an unbiased fashion . Considering one reaction usually contains multiple genes and one gene may be involved in multiple reactions , we combined both gene symbol and KEGG reaction ID to uniquely represent each reaction in the network . We mapped all ZCOGs in larval zebrafish onto zebrafish metabolic network . We obtained 317 ZCOGs coding for enzymes catalyzing 1034 reactions in total . These circadian oscillating enzymes are overrepresented among ZCOGs ( p<0 . 0001 , Fisher’s exact test ) . As circadian phase of ZCOG can be used as the proxy for the peak of circadian activity of enzymes , we defined the distance measure D between reactions incorporating the phase information of ZCOGs together with the proximities of the reactions on the zebrafish metabolic network: Dij=dij+4λsin2 ( pi−pj24 ) π Here dij is the shortest distance between the corresponding reactions i and j on the network with circadian phases pi and pj respectively . λ is a positive parameter to adjust the contribution from phase difference . We then applied a hierarchical clustering of reactions based on these distances . We obtained a main cluster of 632 reactions together with 30 other isolated clusters of much smaller sizes . The tree structure of the main cluster is shown in Fig . 1 when λ = 1 . Five smaller isolated clusters with size larger than 10 are shown in S1 Fig . We defined coherent circadian gene clusters as the clusters obtained from truncating the tree at height with h = 12 so that the genes in each cluster showed coherent circadian phases . The summary of top 10 coherent circadian gene clusters is shown in Table 1 . The genes in the same cluster tend to belong to the closely related metabolic pathways . In particular , we found the top three clusters are all enriched with genes in purine/pyrimidine metabolism . Furthermore , these three clusters showed coherent circadian gene expression around CT2 ( circadian time 2 ) , CT4 , and CT15 respectively , where CT0 is subjective lights-on . Among them , we found that three impdh gene homologs , impdh1a , impdh1b and impdh2 showing strong circadian rhythms of expression fall into two different coherent circadian gene clusters . The peaks of rhythmic expression of impdh2 and impdh1a were observed at CT23 in cluster 1 , while impdh1b exhibited peak expression at around CT12 in cluster 3 . In summary , our method segregated ZCOGs in both metabolic pathways and circadian phases . We next measured the levels of water-soluble metabolites in 5dpf larval zebrafish collected every six hours during 24 hours ( ZT0-ZT24 ) under LD condition using a hydrogen nuclear magnetic resonance ( 1H NMR ) spectroscopy . A total of 173 peak signals were quantified based on their spectral intensities . 73 of them can be assigned to known metabolites including amino acids and nucleotides ( S1 Table ) . We found that 18 of them showed significant circadian oscillation ( cosine fitting p value<0 . 05 and ANOVA p value<0 . 05 ) . These correspond to nine unique known metabolites as shown in Fig . 2 . IMP showed robust circadian rhythm with its peak time around ZT7 . The rest of circadian metabolites include lactate ( ZT9 ) , glutamine ( ZT23 ) , glutamate ( ZT13 ) , threonine ( ZT16 ) , aspartate ( ZT1 ) , creatine ( ZT5 ) , carnitine ( ZT0 ) , and phosphocreatine ( ZT20 ) . We further identified ZCOGs that are linked to these circadian metabolites on our zebrafish metabolic network ( S2 Table ) . Lactate is connected with lactate dehydrogenase Bb ( ldhbb ) peaking at CT17 . Creatine and phosphocreatine are connected with creatine kinase brain b ( ckbb ) peaking at CT4 . Glutamine and glutamate are connected with glutamine synthase , glula and glulb , peaking at CT5 and CT6 respectively . In particular , IMP is directly connected with ten circadian enzymes including impdh homologs in de novo purine synthesis pathway . Therefore , the circadian oscillation of IMP further supports the circadian rhythm of de novo purine synthesis from their enzyme expression . We next investigated the circadian gene expression in adult zebrafish . We monitored the circadian behavior of adult zebrafish ( 6 months old males ) under an infrared behavioral monitoring platform specially designed for adult zebrafish ( S2A Fig . ) . Wild-type ( WT ) zebrafish were raised in 14h:10h light/dark ( LD ) cycle from birth . Adult zebrafish showed robust circadian changes in their locomotor activities under LD ( S2B Fig . ) while the amplitude of oscillation was significantly lower in dark/dark ( DD ) conditions ( S2C Fig . ) . Such strong dependence of circadian activities on light has also been seen in larval zebrafish [38] . The circadian activity in adult zebrafish is more robust than that in larval zebrafish in terms of the baseline activity and the amplitude of the rhythm . We next collected adult zebrafish whole brain every four hours during 48 hours under both LD and DD conditions . The locomotor activity of each adult fish was recorded before being sacrificed ( S3 Fig . ) . The whole-genome transcriptome profiles of adult brain in this time-series were assayed using Agilent zebrafish microarrays . We used a similar statistical method to identify zebrafish circadian genes ( ZCOG ) as our previous study in larval zebrafish . Under a False Discovery Rate ( FDR ) <0 . 05 , we identified 714 ZCOGs in adult brain under both LD and DD conditions ( Fig . 3A ) . 283 ZCOGs were shared by both adult brain and larval ZCOGs ( S2 Dataset and Fig . 3B ) with many exhibiting a similar phase of rhythmic expression during zebrafish post embryonic development ( Fig . 3C ) . In adult brain , impdh1b and impdh2 again showed strong circadian rhythms of expression with impdh1b expression peaking at CT12 , and impdh2 peaking at around CT23 as in larvae . However , impdh1a did not show significant rhythm in adult brain ( Fig . 4A , 4B , 4C ) . We next collected five additional tissues ( eye , intestine , liver , heart and muscle in addition to whole brain ) from adult fish maintained under LD cycles at zeitgeber time 0 ( ZT0 ) ( where ZT0 is defined as lights-on and ZT14 is defined as lights-off ) and ZT12 . Using real-time PCR , we revealed that impdh1a was predominantly expressed in the eye , where the expression level at ZT0 was considerably higher than at ZT12 ( Fig . 4D ) . impdh1b exhibited expression in all tissues , with higher expression at ZT12 than ZT0 in each case ( Fig . 4E ) . Although the expression levels of impdh2 in the heart , muscle and liver were much higher compared with other tissues , the most significant circadian oscillation appeared in the liver ( Fig . 4F ) . These results validated and extended our microarray results . Thus , the circadian expression of three impdh gene homologs are conserved between larval and adult zebrafish . From their spatial and temporal expression patterns , we hypothesize that the three impdh gene homologs have different tissue-specific functions . While impdh1b may play a “housekeeping” role in most tissues , impdh1a is likely to have an eye-related function and impdh2 may function in metabolically active tissues . To explore the functions of the three impdh gene homologs , we knocked down impdh1a , impdh1b and impdh2 expression using morpholino oligonucleotides ( MOs ) respectively . The phenotypes of the three impdh morphants ( MO injected larvae ) at 5 dpf are shown in Fig . 5A-5O . WT zebrafish possess three types of pigment: silver or reflective pigment produced by iridophores ( red arrow in Fig . 5A ) , yellow pigment derived from xanthophores ( yellow arrow in Fig . 5C ) and black pigment derived from melanophores such as that in the retinal pigment epithelium ( RPE ) ( black arrow in Fig . 5C ) . In WT zebrafish embryos , black pigment begins to form around the Prim-5 stage ( 24 hpf ) , xanthophores first appear on the fish head and iridophores are first present on the fish eye at the long-pec stage ( 48 hpf ) [44] . When impdh1a was knocked down , pigmentation defects were visible and easily recognized starting from 72 hpf . At 5 dpf in WT embryos , the iridophore retinal ring is normally filled out and iridophores are densely packed in a dorsal stripe . In contrast , nearly all yellow and silver pigments were absent in impdh1a morphants ( Fig . 5G , I ) and the area covered by melanocytes in the head region was much broader in impdh1a morphants than those in other impdh morphants and controls ( Fig . 5C , F , I , L , O , and Q for quantification ) . In addition to loss of pigment , impdh1a knock-down specifically led to microphthalmia , suggesting that impdh1a plays a crucial role in regulating eye growth . As shown in Fig . 5P , the retina size measured along the anterior to posterior axis was much smaller in impdh1a morphants than in the control groups . Therefore we dissected and sectioned the eyes into slices along the transverse axis to examine in more detail the eye anatomy . In histological imaging of the retina ( Fig . 5B , E , H , K , N ) , none of three impdh homologs morphants showed alterations in the retinal cell types . Furthermore , there were no obvious differences in lens area or pigmentation of the RPE between impdh1a , impdh1b , impdh2 morphants , control and WT except that the overall eye size was much smaller in the impdh1a morphants than in controls . In contrast , both impdh1b and impdh2 morphants displayed normal eye morphology and pigmentation ( Fig . 5 ) . Interestingly , the rate of growth of the impdh1b morphants was higher than controls while impdh2 morphants developed more slowly . Thus at 32 hpf , impdh2 morphants showed a 6–8 h developmental delay ( S4A Fig . ) and only hatched at 52 hpf . In contrast , the more rapid growth of impdh1b morphants was visualized by their longer body length measured between 52 hpf and 5dpf ( S4A , B Fig . ) . Therefore , our data suggest that while impdh2 function tends to increase the rate of zebrafish embryonic development , impdh1b inhibits early growth . Impdh is well known as a rate-limiting enzyme in the de novo biosynthesis of guanine nucleotides . As purines are basic building blocks of DNA and RNA that are needed in cell proliferation , we reason that the circadian rhythm of impdh may influence the rhythm of cell entry into S phase rhythm . To test our hypothesis , we first used MPA , a selective inhibitor of impdh protein , to treat larval zebrafish . We used BrdU to label S-phase nuclei in larvae at six circadian time points as shown in Fig . 6A . In 5 dpf WT larvae , we counted the number of Brdu positive nuclei and found a strong daily S-phase rhythm across 11 somites of the fish trunk ( Fig . 6B ) . We then treated 4 dpf larval zebrafish with MPA at different concentrations and sampled fish starting at ZT0 of 5 dpf at 4h intervals . Larval zebrafish did not show altered morphology or behavioral changes when MPA concentration was below 100μM . MPA treatment significantly reduced the amplitude of the S phase rhythm in a dose dependent manner . This reduction in rhythmic cell proliferation was rescued when 100μM guanosine was added to MPA-treated larval fish ( Fig . 6B ) . These results suggest that MPA inhibits circadian rhythmicity of S phase by limiting the GTP required by DNA synthesis and that de novo purine synthesis plays an important role on the circadian rhythmicity of cell cycle . Next , in order to examine which impdh gene homolog actually contributes to circadian control of cell cycle , we knocked down the three impdh homologs individually using MOs and studied the effect on the circadian rhythm of S phase . We failed to observe any effect of impdh1a MO or impdh1b MO on cell cycle rhythmicity . In contrast , the high-amplitude daily rhythm of S phase was significantly suppressed when impdh2 was knocked down ( Fig . 6C ) . This result suggested that impdh2 plays a major role in circadian control of cell cycle . Impdh has been well documented to serve as a key enzyme in the GTP biosynthesis pathway . In order to explore in more detail the mechanisms whereby changes in purine biosynthesis might generate the overt aspects of the morphant phenotypes , we measured global transcriptome changes in the whole body of impdh1a , impdh1b and impdh2 morphants ( 32hpf ) using RNA-seq . We identified 468 , 331 and 1166 genes whose expression was significantly altered in the impdh1a , impdh1b and impdh2 MO knock-down compared to controls ( S5A Fig . ) . There was only limited overlap amongst these three groups , with only 36 genes showing altered expression in all three MOs ( S5B Fig . ) . This indicates that the down-regulations of three impdh homologs have distinct impacts on global gene expression . We next classified the genes affected by impdh-specific knock-downs into KEGG pathways and examined their tissue specific expression patterns based on annotations in the ZFIN database [38] . The expression of genes affected by the impdh1a knock-down is enriched in retina inner/outer nuclear layers ( S3 Dataset ) while those affected by impdh1b and impdh2 knock-downs are enriched in the regulation of developmental process . Thus , rpe65a involved in visual pigment regeneration and ddt in melanin synthesis are significantly up-regulated upon impdh1a knock-down . Both rpe65a and ddt display circadian rhythms of expression with peaks at CT24 and CT7 respectively . Interestingly , mutations in RPE65 cause retinal pigmentosa similarly to IMPDH1 mutations [45] . dhfr which synthesizes tetrahydrofolate essential for purine metabolism is down-regulated in both impdh1a and impdh1b morphants . In addition , metabolic enzymes including psat1 and phgdh involved in serine synthesis and bcat1 in the synthesis of branched chain amino acids were up-regulated in impdh2 morphants . Both serine and branched chain amino acid metabolism have been implicated as key requirements for cell proliferation [46 , 47] . The differential expression of rpe65a , dhfr , bcat1 and psat1 were all validated by real-time PCR ( S6 Fig . , primers used for real-time PCR are in S3 Table ) . Thus , the molecular functions of the genes affected by three impdh homolog-specific knock-downs provided the clues to the molecular basis of the morphant phenotypes that we observed . In our larval zebrafish microarray data , the expression of genes encoding multiple enzymes upstream of impdh in de novo purine synthesis , including atic , gart , pfas , ppat together with impdh2 , all displayed similar circadian rhythms ( Fig . 7A ) . We further studied their expression in six adult tissues at ZT0 and ZT12 using real-time PCR as shown in S7 Fig . ( primers used for real-time PCR are in S3 Table ) . Similar to impdh2 , all genes showed elevated expression at ZT0 in liver and other tissues . To investigate whether these genes in the purine synthesis pathway were regulated by the core circadian clock mechanism , we generated clock MO knock-down larvae . Time-series measurements of atic , gart , pfas , ppat and impdh2 mRNA levels by real-time PCR showed significantly dampened oscillations in 5 dpf clock morphants compared to WT or control morphants in both LD and DD conditions ( Fig . 7B-7F ) . Thus , these circadian oscillating genes in de novo purine synthesis are likely to be co-regulated by clock . We examined a published Bmal1 ChIP-seq data in mouse liver [48] . In mouse , Ppat and Paics share the same promoter , which contains a strong Bmal1 binding site . Atic also has a Bmal1 binding site in its promoter . We conducted promoter analysis for the corresponding zebrafish genes . We identified a Bmal1/Clock binding site in the promoter of ppat , paics , and impdh2 in zebrafish strongly suggesting that they are also under the direct control of Bmal1/Clock . In mouse retina , ChIP-seq analysis has revealed that Impdh1 is regulated by Crx , a transcription factor specifically expressed in the eye and pineal gland [49] . In our study , crx showed circadian expression peaking at CT17 preceding the peak of impdh1a . The circadian oscillation of impdh1a may be directly regulated by crx and indirectly driven by clock . Our functional analysis implicates impdh1a in melanogenesis and eye development . Our previous study has shown that melanogenesis exhibits a circadian rhythm with a peak around CT0 similar to rhythmic impdh1a expression . Therefore , the circadian oscillation of impdh1a may contribute to circadian rhythm of melanogenesis in zebrafish . In the promoter of impdh1b , we encountered a retinoic acid-related orphan receptor ( ROR ) response element ( RRE ) , the binding site for ROR and NR1D family transcription factors . This result is consistent with the CT12 peak of rhythmic impdh1b expression since we observed previously that circadian oscillating genes with peaks around CT12 tend to contain RRE elements [38] . Taken together , we proposed that the three zebrafish homologs of impdh are regulated by different circadian transcription factors which lead to their different circadian phases and tissue distribution .
Previously , we have demonstrated that a functional circadian clock is already present at 5 dpf in larval zebrafish and we have identified more than 2 , 800 ZCOGs using whole larval RNA extracts and a high-throughput approach [38] . To examine the effects of these ZCOGs on metabolism , we developed a computational method to partition different metabolic processes according to time of day . Our clustering method is based on two assumptions . First , the enzymes similar in circadian phases of gene expression at the transcriptome level also tend to reach their activity peaks approximately at the same time of day . Second , the co-expressed circadian enzymes that are proximal in metabolic network form a module of common metabolic function . Under these assumptions , our method should be generally applicable to the integration of circadian transcriptome data and global metabolic network . Applying our method to zebrafish circadian transcriptome , we found that top clusters of ZCOGs with coherent circadian expression are enriched with genes in purine and pyrimidine metabolism . At a rate-limiting step of this pathway , three impdh gene homologs that have not been previously implicated in circadian clock function , all showed robust , high amplitude circadian rhythms . Furthermore , their circadian expression is developmentally conserved between larval and adult zebrafish . Subsequent functional analysis of these impdh homologs revealed that impdh1a is involved in pigment synthesis and eye development while impdh2 is generally implicated in the control of cell proliferation . Expression of both impdh1a and impdh2 oscillated with circadian peaks at CT23 . In comparison , impdh1b was ubiquitously expressed and showed an opposite phase of rhythmic expression to impdh1a and impdh2 in all tissues tested . Although the detailed functions of impdh1b are still unclear , it seems that it may have the function to delay growth in contrast to both impdh1a and impdh2 which appear to accelerate growth . Our RNA-seq analysis of impdh homolog-specific morphants identified downstream genes that may be responsible for the different phenotypes . Our promoter analysis also suggested that the rhythmic expression of the three impdh homologs was controlled by different circadian transcription factors: impdh2 was regulated by Bmal1/Clock , impdh1a by Crx , and impdh1b by Ror or Nr1d . Fustin et . al ( 2012 ) have reported the circadian expression of genes involved in purine and pyrimidine nucleotide metabolism in mouse liver [50] . They revealed that circadian rhythms of Ppat and Pnp expression contribute to rhythmic synthesis of guanine nucleotides but they fell short of linking this phenomenon to circadian control of the cell cycle . In mouse , Ppat shows strong circadian oscillation in multiple tissues with a peak at CT11 , typical for Bmal1/Clock regulated circadian genes . Consistently , our previous larval study identified rhythmic expression of several pnp homologs in zebrafish [38] . Our results showed that both MPA inhibition and impdh2 knockdown have significantly dampened the amplitude of circadian rhythm of S phase in proliferating cells of zebrafish . This is consistent with the earlier report that MPA and impdh2 have the same inhibitory effect on the sprouting of intersegmental blood vessels in early zebrafish larvae [51] . We found that IMP , the substrate for Impdh2 , is peaking at ZT7 , five hours ahead of the peak of S phase . This observation suggests that the peak of Impdh2 activity may be between ZT7 and ZT12 to deplete IMP level to provide purine for DNA synthesis . Thus the circadian oscillation of IMP together with impdh2 can have an important regulatory role on the metabolic flux through purine synthesis pathway . From the microarray data for mouse tissues and cell lines in the BioGPS database ( http://biogps . org/ ) , mouse Impdh2 together with other de novo purine synthesis genes are highly expressed in embryonic stem cells . From the in situ data for the developing mouse embryo at E14 . 5 in Metscout database ( http://www . metscout . mpg . de ) , Impdh2 is expressed in proliferating cells including those in the ventricular zone of the brain similar to other enzymes involved in de novo purine synthesis . It was shown that enzymes in de novo purine synthesis form reversible multi-enzyme complex in so-called “purinosome” in HeLa cells [52] . These observations suggest that the enzymes in de novo purine synthesis pathway are co-regulated both spatially and temporally to control metabolic flux . Karpowicz et al . ( 2013 ) have found that the circadian clock regulates stem cell regeneration in the fly intestine [53] . In mammals , there are also reports of circadian clock regulation in epidermal cells [54] and hematopoietic stem cells [55] . Therefore , our results suggest that circadian purine synthesis may directly contribute to cell proliferation and stem cell regeneration regulation by the circadian clock in a highly conserved manner . Our study has revealed that impdh1a is required for pigment synthesis and ocular development . Interestingly , the gart and paics mutants which affect two enzymes upstream of Impdh in de novo purine synthesis , display almost identical phenotypes in the zebrafish retina as our impdh1a morphant [56] . From this previous study it was proposed that purine synthesis provides key precursors for three types of pigment cells in zebrafish: iridophores , xanthophores and melanophores [56] . Our results suggest that impdh1a is also involved in these processes . This is consistent with our previous finding that melanogenesis is under circadian clock control in zebrafish [38] . Therefore , it seems that impdh1a and impdh2 along with other enzymes in de novo purine synthesis are involved in circadian regulation of two seemingly different processes that are both controlled by purine synthesis: pigment synthesis in pigment cells and cell growth in proliferating cells . This functional difference is also revealed by the distinct differential gene expression upon their MO knockdowns . It is well known that circadian clock dysfunction is closely associated with metabolic disorders or diseases . Metabolism-based research in circadian rhythms represents a potentially important contribution to our understanding of circadian output functions [57] . The mis-regulation of enzymes in nucleotide metabolism is especially prone to diseases . Only two IMPDH homologs , IMPDH1 and IMPDH2 , exist in humans sharing 84% amino acid identity [58] . The amino acid sequences of Impdh homologs are highly conserved between human , mouse and zebrafish [51] . Thus , zebrafish Impdh1a shares 90% identity and Impdh1b , 91% identity with human IMPDH1 ( S8 Fig . ) . In addition , zebrafish Impdh2 shares 91% identity with human IMPDH2 . Among the two IMPDH homologs that have been described in humans , IMPDH1 is linked to several types of severe retinal degeneration including retinitis pigmentosa and Leber congenital amaurosis ( LCA ) [45] . Furthermore , IMPDH2 contributes to cell proliferation in a variety of cell lines and tissues such as activated lymphocytes and tumor cells [59–61] . Thus , IMPDH has already become an important drug target for antiviral , cancer chemotherapy and immunosuppressive therapy . Immediately downstream of IMPDH , hypoxanthine phosphoribosyltransferase ( HPRT ) that also showed circadian expression similar to IMPDH2 in our study is implicated in Lesch-Nyhan syndrome , a neurological disease [62] . Therefore , our study provides novel insight into whether enzymes in de novo purine synthesis , especially IMPDH , can be used as biomarkers for metabolic diseases which are linked with circadian clock disorders .
Zebrafish handling procedures were approved by the Institute of Neuroscience , Shanghai Institutes for Biological Sciences , Chinese Academy of Sciences . Wild type AB strain zebrafish ( Danio rerio ) were obtained from National Zebrafish Resources of China ( Shanghai Institutes for Biological Sciences ) . Adult zebrafish ( 6 months old , male ) and larval zebrafish were maintained at 28°C under a 14h: 10h light/dark cycle . The light provided by three full spectrum fluorescent bulbs ( 13W , SHLY ) was turned on at 9:00 , and turned off at 23:00 . The illuminance during light exposure was approximately 1400 lux , measured by a digital luxmeter ( Model ZDS-10 , SHXL ) at the water surface . The fish were fed live brine shrimp at 10:00 and 15:30 every day . We downloaded all enzymatic reactions and their associated genes , i . e . Gene-Protein-Reaction ( GPR ) association , for zebrafish included in the KEGG database . We obtained the major metabolites participating in these reactions and the direction of reactions by parsing the KEGG pathway maps . The information of the zebrafish metabolic network can be found in S1 and S4 Datasets . The reactions are considered to be connected if they share the same substrate or product . This leads to an un-directed graph with reactions as nodes . The shortest distances between reactions on the graph were computed using Floyd’s algorithm implemented by allShortestPaths function of e1071 library in R program . The distances between genes coding for the enzymes catalyzing the reactions were computed as the sum of the shortest distances between their associated reactions ( if they are associated with the same reaction ) and the contribution from their phase difference in circadian gene expression . The contribution to the distance from phase difference is proportional to 4sin2 ( Δθ2 ) , the squared distance between two unit vectors on the complex plane with phase angle difference Δθ . The phase angle of the unit vector is related to the peak time of circadian oscillating gene between 0 and 24 by θ=p×2π24 . The term 4sin2 ( pi−pj24 ) π gives rise to the equivalent distance of 0 , 1 , 2 , 3 , 4 when the peak times of two genes differ by 0h , 4h , 6h , 8h , 12h respectively . Based on the combined distances in metabolic network and peak time difference , the gene tree was constructed by hierarchical clustering with complete linkage in R program . The tree structures in Fig . 1 and S1 Fig . were visualized using iTOL [42 , 43] , and were not drawn to scale . To examine genome-wide circadian gene expression , we sampled adult male zebrafish whole brains in both LD ( 14h:10h light/dark ) and DD conditions for microarray . In the LD group , 24 fish were placed in an adult zebrafish activity recording system ( S1 Text ) . After 3 days 14h:10h LD acclimation , the fish were sacrificed and dissected at 4h intervals starting at ZT0 of the fourth LD cycle for 12 time points . In the DD group , 24 fish were also acclimated to 14h:10h LD cycle for 3 days , and then they were sacrificed every 4h starting at CT0 of the fourth day in DD condition for 12 time points . 2 fish were sampled individually in each time point . Every fish was killed by rapid decapitation , and the whole brain was removed and frozen immediately in liquid nitrogen and stored at -80°C . The collection of samples under darkness was performed under dim red light . Locomotor activity of each zebrafish was recorded during the experiment . Microarray data analysis was performed as described before [38] with slight modifications ( S1 Text ) . Briefly , the selection criteria used here were as follows: g-test p values less than 0 . 3 in both LD and DD with a dominant period set as 24 h . These cutoffs corresponded to an overall FDR less than 5% as computed from random permutation . The microarray data have been deposited in Gene Expression Omnibus ( GEO ) under accession number: GSE51279 . Real-time PCR was performed as previously described [38] . The primer sequences of genes tested are listed in S3 Table . Morpholinos ( MOs ) were purchased from Gene Tools . Each MO except standard controls was designed to target the start codon region of the gene . The sequence of the clock MO was 5’-CAT CCC GGT CTA TGC TGG AGG TCA T-3’ as previously used by Li et al . [63] . The sequence for impdh1a MO was 5’-GAT CAG GTA ATC AGC CAT GAG TCT C-3’; impdh1b MO , 5’-CTC CGC TTA TCA GAT AGT CTG CCA T-3’; impdh2 MO , 5’-GCT GAT TAA ATA GTC CGC CAT AGT-3’; and the standard control MO , 5’-CCT CTT ACC TCA GTT ACA ATT TAT A-3’ . The sequences of impdh1a MO and impdh2 MO were the same as those used previously [51] . MOs were used at the following doses: clock MO: 2 . 5ng; impdh1a MO , 8 ng; impdh1b MO , 9 . 6 ng; impdh2 MO , 5 . 6 ng; standard control MO for clock knock-down , 2 . 5ng; standard control MO for three impdhs knock-down , 9 . 6ng . MOs were pressure-injected into 1- to 2-cell stage embryos at a volume of 1 nl using Picospritzer II injectors as described [64] . Mycophenolic acid ( MPA , Sigma ) was dissolved in dimethylsulfoxide ( DMSO ) at a stock concentration of 100mM and guanosine ( Sigma ) was also dissolved in DMSO at a stock concentration of 1M . At 4 dpf , WT larvae raised on a 14h: 10h LD cycle at 28°C were distributed into 35mm dishes with each dish holding 25 larvae in 5ml E3 buffer . Larvae were then exposed to drugs beginning at ZT0 ( 96 hpf ) by directly pipetting MPA and/or guanosine stock solutions into each dish at the following final concentrations: 1μM , 10μΜ , 100μM; MPA and 100μM guanosine . The DMSO final concentration was adjusted to 1% in each dish . At their final concentrations , MPA and guanosine had no detectable effect on larval zebrafish behavior . Drug treated larvae were collected at 4h intervals starting at ZT0 of 5 dpf for 6 time points in LD conditions . 5 dpf larvae were incubated in 15% DMSO and 85% hanks ( Gibco ) solution with 10 mM bromodeoxyuridine ( BrdU , Sigma ) for 20min on ice . Then larvae were fixed in 4% paraformaldehyde ( PFA ) at 4°C overnight . After gradual dehydration and rehydration in methanol , larvae were permeabilized 10 min with 100% acetone prechilled to −20°C , and then 20 min with 0 . 1% trisodium citrate ( with 0 . 1% Triton X-100 ) at RT . Larvae were rinsed three times with ddH2O and incubated in 2N HCl for 1h at RT . Subsequently larvae were washed in PBST ( 0 . 1% Triton X-100 in PBS ) several times and blocked in 2% goat serum for 2h at RT . Then whole-mount immunofluorescence was performed with 1:10 anti-BrdU antibody ( Roche ) at 4°C overnight followed by incubation with secondary antibodies conjugated to fluorescein ( 1:1000 ) according to standard protocols . Brdu labeled larvae were embedded in 1 . 0% low-melting point agarose for imaging . Images were performed on an Olympus FV1000-MPE laser scanning confocal microscope . The number of Brdu positive nuclei was counted over 11 somites of the fish trunk between the anus and head . Melanocyte area measurement was performed as described before [38] . The covered area of melanocytes was measured in the head region from the pineal gland to the optic vesicles excluding the eyes . Zebrafish eye histological analysis was conducted on vibratome sections . 5 dpf larvae were collected and fixed overnight at 4°C in 4% paraformaldehyde ( PFA ) . After washing several times in PBST ( 0 . 1% Triton X-100 in PBS ) , the whole larvae were embedded in 4% low-melting point agarose and 50-μm-thick sections were cut using Leica VT 1200S . Images were obtained on an Olympus microscope ( 1X71 ) mounted with a DP72 digital camera controlled by DP2-BSW software . To maximize comparability across different impdh morphants , only sections with an optic nerve visible in the eyes were used . 5 dpf larvae morphological images were taken using an Olympus microscope SZX16 equipped with a DP71 CCD camera controlled by DP Controller software . Larval zebrafish anterior-posterior eye length and body length ( from the middle of the mouth to the tip of the tail ) was measured on images using ImageJ 1 . 41 software . Images were processed using Adobe Photoshop CS2 . To validate the functions of the three impdh genes , we collected impdh MO injected embryos for RNA-seq . WT larvae , control and impdh morphants were raised under LD conditions and sampled simultaneously at 32 hpf . Each group had at least 40 embryos . Total RNA was extracted from each sample using Trizol . The quantity and quality of the RNA samples were assessed with a Qbit ( invitrogen ) and an Agilent 2100 bioanalyzer . RNA-seq was performed by Partner Institute for Computational Biology Omics Core . mRNA-seq libraries were prepared using the TruSeq RNA Sample Preparation v2 Kit following the manufacturer's protocol ( Illumina ) . Briefly , 1ug poly-A containing mRNA was purified using poly-T oligo attached magnetic beads and then fragmented . The cleaved RNA fragments were primed with random hexamers into first strand cDNA , and then the second-strand cDNA was synthesized . The ends of double-strand cDNA ( ds cDNA ) were repaired into blunt ends , after which an adenosine base was added to the 3’ ends of ds cDNA and adaptors with a single T base overhang on 3’ end were ligated . These adapter-modified cDNA fragments were amplified by PCR which was performed with a PCR primer cocktail that annealed to the ends of the adapters . The products were purified , then the concentration and size distribution of the libraries were determined on a Qbit and an Agilent Bioanalyzer . Libraries were loaded onto single flow cells at concentrations of 9 pM to generate cluster densities of 750 , 000–850 , 000/mm2 following Illumina's standard protocol using the Illumina cBot and TruSeq SR cluster kit V3-HS ( cBot ) . The flow cells were sequenced as 51 single reads on an Illumina HiSeq 2000 using TruSeq SBS sequencing kit version 3 and HCS version 2 . 0 data collection software , respectively . Base calling was performed using Illumina's real time analysis ( RTA ) . The row sequence reads were exported in FASTQ format . Sequence reads from RNA-seq data were aligned to the Danio rerio genome ( Zv9 ) using the bowtie2 program [65] and all read mapping was carried out using TopHat2 [66] . All expression values were estimated using Cufflinks 2 . 1 . 1 ( http://cufflinks . cbcb . umd . edu ) [67] with the same annotations and reference sequences as TopHat2 . FPKM value was obtained for each predicted transcript . We excluded the low abundant transcripts with FPKM less than 1 in all samples . Differential gene expression between each pair of samples was characterized by the log2 ratio of their FPKM values . A gene was considered as significantly differentially expressed if the absolute value of log2 ratio of FPKM values was greater than 2 . The RNA-seq data have also been deposited in GEO under accession: GSE51279 . WT AB larvae were raised in 14h:10h LD condition and collected at 5dpf . The larvae were sampled at 6h-interval: ZT0 , ZT6 , ZT12 , ZT18 , ZT24 , and every time point has three biological replicates . For each sample , 50 larvae were collected in tubes for homogenization ( 91-PCS-CK14 , PeqLab ) and rapidly frozen in liquid nitrogen . Considering ideal freezing and storage conditions for pH-sensitive samples [68] , the frozen samples were freeze dried overnight . Ceramic beads ( 91-PCS-CK14 , PeqLab ) and 1000 μL of acetonitrile / H2O ( 1:1 ) were added and the samples were extracted for 25 minutes at 4°C using a neolab Intellimixer ( u1 , 99 rpm ) . The samples were centrifuged at 14 , 000 rpm for 10 minutes at 4°C and the supernatants were freeze dried . For 1H-NMR measurement , the extracts were dissolved in a mixture of buffer ( 1 . 5 M KH2PO4 , 2 mM NaN3 , 0 . 1% TSP in D2O , pH 7 . 1 ) and D2O ( 1:9 ) . Spectra were recorded on a Bruker Avance II spectrometer using a 1H-BBI double resonance probe ( Bruker Biospin GmbH ) . 1D NOESY spectra were recorded with presaturation for water suppression and 256 scans at 300 K ( 26 . 9°C ) . A prescan delay of 4 s was used together with a mixing time of 10 ms . Pulse lengths were determined automatically by the Bruker AU program pulsecal . 64k complex data points corresponding to a sweep width of 12 , 345 . 68 Hz were recorded . All spectra were treated identically using an exponential apodization function , introducing an additional linewidth of 0 . 3 Hz and automated phasing , baseline correction and referencing using the Bruker macro apk0 . noe . Preliminary peak assignment was done using databases ( Chenomx ( Chenomx Inc . ) , BBIOREFCODE ( Bruker BioSpin GmbH ) , The Human Metabolome Database ( HMDB ) ) and confirmed by spiking pure substances in the samples . Integration was done using AMIX ( Bruker Biospin GmbH ) . Spectra were scaled to total intensity . Circadian oscillations were identified by fitting the mean intensities across 24 hours to a cosine function with periods between 20–28 and shifting phases . Statistical significance of differences between groups collected at different time points were assessed by ANOVA . | The circadian timing system gates cell cycle progression in various organisms from unicellular microorganisms to mammals . Although a number of factors have been implicated in linking the circadian clock with the cell cycle , little is known about the contribution of metabolic processes to this interaction . In this study , we have shown that cell cycle , metabolism and the circadian clock are intertwined through consistent circadian expression of genes in de novo purine synthesis in the zebrafish . Three gene homologs of a key enzyme in this pathway , impdh , show circadian oscillations in different tissues and have distinct molecular functions . In particular , one of the homologs , impdh2 , contributes to the daily rhythm of S phase in the cell cycle . We suggest that de novo purine synthesis mediates circadian control of cell cycle , a mechanism that is likely conserved in mammals . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [] | 2015 | Integrative Analysis of Circadian Transcriptome and Metabolic Network Reveals the Role of De Novo Purine Synthesis in Circadian Control of Cell Cycle |
Recent reports of transmission interruption of Onchocerca volvulus , the causing agent of river blindness , in former endemic foci in the Americas , and more recently in West and East Africa , raise the question whether elimination of this debilitating disease is underway after long-term treatment of the population at risk with ivermectin . The situation in Central Africa has not yet been clearly assessed . Entomologic data from two former endemic river basins in North Cameroon were generated over a period of 43 and 48 months to follow-up transmission levels in areas under prolonged ivermectin control . Moreover , epidemiologic parameters of animal-borne Onchocerca spp . transmitted by the same local black fly vectors of the Simulium damnosum complex were recorded and their impact on O . volvulus transmission success evaluated . With mitochondrial DNA markers we unambiguously confirmed the presence of infective O . volvulus larvae in vectors from the Sudan savannah region ( mean Annual Transmission Potential 2009–2012: 98 , range 47–221 ) , but not from the Adamawa highland region . Transmission rates of O . ochengi , a parasite of Zebu cattle , were high in both foci . The high cattle livestock density in conjunction with the high transmission rates of the bovine filaria O . ochengi prevents the transmission of O . volvulus on the Adamawa plateau , whereas transmission in a former hyperendemic focus was markedly reduced , but not completely interrupted after 25 years of ivermectin control . This study may be helpful to gauge the impact of the presence of animal-filariae for O . volvulus transmission in terms of the growing human and livestock populations in sub-Saharan countries .
The interruption of transmission of Onchocerca volvulus , the causing agent of river blindness or onchocerciasis , has been confirmed for a growing number of endemic foci on the American continent [1 , 2 , 3] and in West [4] and East Africa [5 , 6 , 7] . The recent success in onchocerciasis control can be mainly attributed to the extensive and sustained mass treatment programs with the microfilaricide ivermectin , governed by institutions of the World Health Organization , like the African Programme for Onchocerciasis Control [8] . Long treatment rounds are necessary because the drug is only lethal to the larval stage and not the adult worm . There are thus good prospects that elimination of onchocerciasis is well underway in the Americas [9] and may also have begun at different foci on the African continent [6 , 10 , 11] . However , currently there is a paucity of information on the actual situation in Central and Southern Africa , in particular with respect to vector transmission , albeit a significant proportion of these regions have been hyperendemic . Recent studies on the effects of ivermectin treatment on the epidemiology of O . volvulus in humans and the black fly vector Simulium damnosum sensu lato have been done in North and West Cameroon [12 , 13 , 14 , 15 , 16] . The caveat of the most recent studies is that the filarial species in the vector were not always correctly identified , and the prevalence of infective O . volvulus larvae and thus the transmission potential remains unknown . Local S . damnosum s . l . populations are vectors of at least two other species from the Onchocerca genus: O . ochengi , a common parasite of Zebu cattle Bos indicus [17] and O . ramachandrini , a filaria from the warthog Phacochoerus africanus [18] . The proportion of animal-filariae in the vector has direct and indirect consequences for parasite transmission to humans [19 , 20 , 21] rendering it an important factor to understand the epidemiology of river blindness . Furthermore , filariae closely-related to O . volvulus might repopulate the human host [22] posing a potential risk of infection , or they might transfer genes to O . volvulus which negatively affect the effectiveness of ivermectin , presently the sole drug intervention in use [23] . For this reason we combine microscopic differentiation of infective larvae with a PCR-based molecular approach which allows the detection of yet unknown filarial species and strains in addition to already known Onchocerca spp . [22] . This study presents the latest entomologic data of a longitudinal study in the Vina du Nord valley , North Cameroon , dating back to 1976 when ivermectin mass treatment had not yet commenced [24 , 25] . The impact on O . volvulus transmission after 25 years of annual community-directed treatment with ivermectin ( CDTI ) is demonstrated here . Furthermore , a second site endemic for onchocerciasis in an economically important cattle livestock region has been monitored since 1985 . The epidemiologic data is also complemented with Onchocerca spp . transmitted by the same local vectors of the S . damnosum complex and discussed in light of their impact on transmission success of O . volvulus . We have not studied onchocerciasis transmission in regions where ivermectin treatment is contraindicated , such as co-endemic foci of Loa loa in the Central African rainforest , although they remain potential source-areas for reinvasion .
Two S . damnosum fly catching sites at two foci in Northern Cameroon were visited between two to four times per month ( Fig 1 ) . One former hyperendemic onchocerciasis focus is the village Soramboum close to the Vina du Nord river in the Sudan savannah ( 500 m altitude ) : 7°47'14"N; 15°0'22"E where ivermectin mass treatments have been conducted since 1987 . Here we present entomological data collected from September 2009 till March 2013 . The village has approximately 1000 inhabitants today , and between 1000 and 2000 cattle are located in the vicinity ( personal observation ) . The Vina du Nord river is flowing perennially with an average annual water discharge of 150 m3 per second , with highest values between July and October [26] . The other formerly hypo- to mesoendemic focus monitored is the village Galim located 15 km south of Ngaoundéré ( population: 500 inhabitants , approximately 5000 cattle in the vicinity , personal observation ) at the Vina du Sud river ( mean annual water discharge: 37 m3/s , 1050 m altitude ) : 7°12'2"N; 13°34'56"E , where CDTI has started in 1997 . The entomological data was collected from April 2009 till March 2013 . The area belongs to the Guinea-grassland of the Adamawa plateau , located in an important area for cattle livestock production in Cameroon [27] . Baseline and follow-up data of O . volvulus transmission to man before ivermectin mass treatments started is available for both foci [15 , 16 , 19 , 25 , 28] , and publicly available via the project website www . riverblindness . eu ( http://riverblindness . eu/epidemiology/fly-catching-sites-data/ ) . Whereas S . damnosum sensu stricto and S . sirbanum are the predominant vector species at the Vina du Nord river , it is S . squamosum at the Vina du Sud river [29] . Fly catching was performed according to Duke et al . [30] with the following modifications . Blood meal-searching female Simulium flies were attracted on man by exposing the fly catcher’s legs and trapped with sucking tubes as soon as they settled before starting to probe . Usually the catching period was from 7 am to 6 pm , interrupted by a one hour break at noon . Afterwards , the catches were transported to the research station in Ngaoundéré and stored at -15°C until dissection . Flies were sorted , counted , and a maximum of 100 female S . damnosum s . l . flies per site and day were dissected with needles under a stereomicroscope ( Wild M5 , Switzerland ) . The parous rate was determined by examination of the ovaries in the abdomen [31] . From parous flies infested with filarial worms the location ( head , thorax or abdomen ) , molting stage and quantity was noted . Following the identification key of Wahl et al . [15] , infective third-stage larvae ( L3 ) were classified to species according to body length , measured by an ocular eye-micrometer attached to the stereomicroscope at 50x magnification , and shape of the anterior and posterior ends . For a subdivision of the L3 taken between February 2010 and February 2012 , a molecular investigation of their mitochondrial DNA was conducted according to Eisenbarth et al . [22] . Briefly , single L3’s were digested with 1 to 2 μl proteinase K ( 20 μg/μl stock ) in 75 μl DirectPCR lysis reagent ( Viagen Biotech , USA ) at 55°C . Two μl of the crude extract was used for each 25 μl PCR reaction . Primer pairs of three mitochondrial loci ( 12S rRNA , 16S rRNA and coxI ) that allow for the discrimination of filarial species were used . The amplified PCR products were sequenced on an ABI Prism 3100 genetic analyzer ( Applied Biosystems , USA ) following the manufacturer’s protocol . For the comparison of the body lengths of L3's identified by molecular markers , a larger sample size was taken from flies caught in the same period at the Vina du Sud river about one kilometer downstream of the site near Galim . These flies were collected both from man and cattle . The Annual Biting Rate ( ABR ) and Annual Transmission Potential ( ATP ) of Simulium damnosum flies were determined according to the literature [24 , 25 , 32] . First , the monthly biting rates ( MBR ) were calculated by the sum of the flies caught per month divided by the number of catching days and multiplied by the number of days per month . No correction was made for the missing hours due to rain , sandstorm , or any other reasons . The ABR is the sum of 12 MBRs per hydrological year , measured from April ( beginning of rainy season ) till March next year ( end of dry season ) . For months during which no catches were attempted , the mean MBR value for the corresponding month and site over the respective decade was estimated by interpolation . The monthly infection rate was the sum of the infective L3 of O . volvulus , O . ochengi and O . ramachandrini from the head , thorax and abdomen found in all parous flies , divided by the sum of dissected flies . By multiplying the monthly infection rate with the respective MBR , the Monthly Transmission Potential ( MTP ) was determined . The ATP is the sum of 12 MTPs for one year , starting from April till March next year . Missing data points were extrapolated by the sum of all MTP divided by the number of months with data , and multiplied by factor 12 . If the MTP data per year was below 3 , the mean annual infection rate of proximate years multiplied with the respective ABR was used instead for the ATP calculation . The statistical program Python version 3 . 4 . 1 was used for statistical analysis employing student t-tests . Results were considered statistically significant when the p-value was below 0 . 05 . P-values were corrected for multiple testing by multiplying with the number of tests done . The effect size was calculated according to Cohen [33] . For depicting the distribution of the L3 body length from a random sample , violin plots , i . e . box plots with a rotated kernel density plot on each side , were used .
During the study period , a total of 39 , 082 flies were caught on human fly catchers , and 21 , 897 ( 55 . 6% ) of them were dissected: a total of 2096 L3 were found ( Table 1 ) . Depending on the catching site , a mean of 1 . 96 ±0 . 53 L3 ( max . 20 ) were harvested per infective fly in Soramboum and a mean of 3 . 49 ±0 . 13 L3 ( max . 23 ) in Galim . Near Soramboum at the Vina du Nord site the infection rate ( flies carrying L1 , L2 and L3 ) of parous flies in the rainy season was significantly higher than in the dry season ( mean: 10 . 9 vs . 5 . 9% , t-value = -4 . 91 , p < 0 . 001 ) , as well as the infection rate with infective L3 stages ( mean: 7 . 8 vs . 3 . 3% , t-value = -4 . 88 , p < 0 . 001 ) . The opposite was true at the Vina du Sud site near Galim ( mean infection rate: 11 . 1 vs . 16 . 6% , t-value = 2 . 36 , p > 0 . 05; mean L3 infection rate: 4 . 3 vs . 7 . 1% , t-value = 1 . 89 , p > 0 . 05 ) . Moreover , the parous rate , which is a parameter of age structure , was on average 9 . 6% higher in the wet season than in the dry season in Soramboum ( range: -18 . 3–19 . 3% , t-value = 5 . 31 , p < 0 . 001 ) , but in Galim on average 11 . 8% lower in the wet season compared to the dry season ( range: 6 . 7–24 . 3% , t-value = 3 . 83 , p < 0 . 001 ) . The higher proportion of infective L3 to developing larvae during the rainy season in Soramboum ( 64 . 0% vs . 53 . 5% , t-value = -1 . 88 , p > 0 . 05 ) , but not in Galim ( 33 . 9% vs . 35 . 3% , t-value = 0 . 88 , p > 0 . 05 ) is lacking statistical support . A generally higher proportion of L3 in Soramboum can be explained by a longer storage time of the caught flies at ambient temperature during the time ( often days ) until they were brought to the laboratory , 225 km away by public transport , so that more larval stages developed further . Fig 2 shows the ABR for the two study sites starting prior to the distribution of ivermectin . With the exception of 1995 the ABR in Galim was higher than in Soramboum , on average by a factor of 4 . 26 ( SD ±3 . 30; range: 0 . 42–14 . 67; n = 17 ) . The yearly fluctuations were more pronounced in the Vina du Nord valley and followed a cyclical pattern ( Fig 2A ) . In contrast , the ABR at the Vina du Sud fluctuated only mildly apart from intermittent dips , which reached previous levels in the following year ( Fig 2B ) . An ongoing trend of lower biting frequencies was evident in Soramboum since 2002 ( mean: 19 , 700 flies per person and year vs . 35 , 348 before ) and in Galim since 2006 ( mean: 39 , 628 flies per person and year vs . 103 , 564 before ) . In Soramboum the decline in biting rate occurred mainly in the dry season from October till March with only little changes during the rainy season ( Fig 3A ) , whereas in Galim the highest decline was within the peak of the dry season and the peak of the rainy season from February till August ( Fig 3B ) . In the same period of declining ABRs , the monthly infection rate of all L3-harboring flies of all Onchocerca spp . increased in Soramboum from 2 . 25% ( 1987–2001: 95%-CI: 0 . 45; n = 90 ) to 3 . 26% ( 2002–2012: 95%-CI: 0 . 53; n = 89 ) , while it remained stable in Galim ( 1989–2005: 3 . 34% , 95%-CI: 0 . 57 , n = 100 vs . 2006–2012: 3 . 19% , 95%-CI: 0 . 62 , n = 65 ) . A historic summary of the Annual Transmission Potentials over the last 36 years in Soramboum ( Fig 4A ) and 27 years in Galim ( Fig 4B ) illustrates the alterations in the ratio of animal-filariae and the human filaria O . volvulus in the vector . In Galim , annual filarial transmission rates remained high till 2006 ( mean: 13 , 525 L3 per person and year , SD ±5334 ) , when it dropped to 32 . 5% of previous levels ( mean: 4395 L3 per person and year , SD ±2348; Fig 3B ) . In contrast , Soramboum experienced an increase of L3 transmission after the early years of ivermectin mass treatments , from an average ATP of 1045 ±438 L3 per person and year in 1987–88 to 2286 ±1338 in 1993–98 , which later returned to former levels , i . e . an ATP of 1242 ±741 in 1999–2012 ( Fig 4A ) , although the pre-ivermectin control ATP from the adjacent Touboro site was much higher ( 4140 L3 per person and year , Fig 3A ) . According to morphological classification the species composition of the L3 population in Soramboum from 2009 till 2012 was 23 . 9% O . volvulus , 65 . 9% O . ochengi and 10 . 2% O . ramachandrini ( Fig 5A ) . In previous years the species composition of O . volvulus—O . ochengi—O . ramachandrini fluctuated from 60 . 7%–12 . 3%–27 . 0% ( 1987–88 ) to 22 . 3%–65 . 3%–10 . 2% ( 1993–99 ) and 40 . 5%–50 . 8%–8 . 7% ( 2000–06 , Fig 4A ) . Correspondingly , the species were composed as follows in Galim: 11 . 3% O . volvulus , 88 . 7% O . ochengi ( 1995–96 ) , 19 . 2% , 80 . 8% ( 2000–05 ) and 17 . 4% , 82 . 6% ( 2006–12 , Fig 4B ) . No O . ramachandrini L3 were not found at all . At the Vina du Nord site 96 isolated L3 ( 10 . 3% of all found ) from 52 infected flies ( 7 . 2% of all dissected ) were subjected to molecular identification , of which 76 ( 79 . 2% ) PCR products were successfully sequenced . At Galim from the Vina du Sud site , 40 L3 ( 3 . 5% of all found ) from 22 infected flies ( 2 . 8% of all dissected ) provided 28 ( 70% ) amplicons of Onchocerca spec . which could be successfully sequenced . The species composition of these L3 from Soramboum was 6 . 6% O . volvulus , 76 . 3% O . ochengi and 17 . 1% O . ramachandrini ( Fig 5B ) , whereas in Galim only O . ochengi was found ( Fig 5D ) . A recently discovered O . ochengi genotype called 'Siisa' [22 , 23 , 34] contributed to 8 . 6% and 10 . 7% , respectively , of the local O . ochengi L3 population in Soramboum and Galim ( Fig 4B and 4D , respectively ) . In comparison with morphological classification ( n = 71 ) , 72 . 2% ( 13/18 ) of so-called O . volvulus in Soramboum were in fact O . ochengi , and 2 . 4% ( 1/41 ) of O . ochengi were O . ramachandrini . All examined O . ramachandrini ( n = 12 ) were correctly identified . Hence , the respective effective ATP in Soramboum for the years 2009 to 2012 was 68 , 221 , 58 and 47 for O . volvulus ( mean: 98 ) : 773 , 2503 , 1388 and 475 for O . ochengi ( mean: 1285 ) and 18 , 392 , 238 and 70 for O . ramachandrini ( mean: 180 ) . Accordingly , the adjusted species proportion of the L3 population for these years were on average 6 . 3% O . volvulus , 82 . 2% O . ochengi and 11 . 5% O . ramachandrini ( Fig 4A , right side ) . At the Vina du Sud site near Galim ( n = 67 ) O . volvulus ( 0/46 ) and O . ramachandrini ( 0/0 ) have not been detected since the introduction of molecular methods for L3 species identification ( Fig 5D ) , although there were morphologically identified specimens of O . volvulus ( n = 149 , 14 . 6% of total , Fig 5C ) . All morphologically classed O . ochengi ( n = 21 ) were correctly identified . Hence , the whole L3 population in the observation period 2009–2012 consisted of O . ochengi ( Fig 4B , right side ) with an ATP of 5096 , 4525 , 6753 and 2046 ( mean: 4605 ) . In order to evaluate the reliability of body length as a characteristic trait that can be used for species discrimination of infective larvae , the body lengths of occurring Onchocerca spp . L3 in S . damnosum s . l . were compared with morphological and molecular identification methods ( Fig 6 ) . Whereas the inter-specific differences according to morphological criteria are significant ( p < 0 . 001 ) , no within-species length difference has been detected between morphological and molecular identification of O . volvulus and O . ramachandrini . A significant ( p < 0 . 01 ) within-species difference has been found in the common genotype of O . ochengi sensu stricto , but with a low effect size ( d = 0 . 598 , n = 95 ) ; a significant difference ( p < 0 . 001 , d = 1 . 471 , n = 19 ) was also found for the genotype O . ochengi 'Siisa' . Interestingly , a more than 4-times higher proportion of morphologically misidentified O . volvulus were O . ochengi 'Siisa' ( 25 . 4%; 15/59 ) than in the morphologically identified O . ochengi group ( 6 . 3%; 4/64 ) . For the genotype O . ochengi s . s . , this was vice versa ( 74 . 6%; 44/59 of misidentified O . volvulus vs . 92 . 2%; 59/64 of O . ochengi group ) .
Since 2006 there is a steady trend of lower ABRs , in particular at the Vina du Sud river , where biting rates before were with only one exception above 60 , 000 per man and year ( Fig 2B ) . The vector transmission of filarial stages have also dropped during this time ( Fig 4B ) , but to a lesser extent in the Sudan savannah ( Fig 4A ) due to a concomitant gain of the infection rate by bovine filariae . The reason for this vector decline could be the result of decreased availability of breeding sites or food for the aquatic Simulium larvae , and hence a drop in population size . A distinct increase of endoparasitic mermithids in human-biting nulliparous flies was evident at the Vina du Sud breeding sites ( S1 table ) . It is , however , unlikely that these mermithids or other Simulium parasites are the main drivers for the massive decline in biting rates of recent years . A reduced longevity of adult flies was not observed , when comparing the current parous rate with those of flies at baseline [24] . Furthermore , a continuous rise in the pool of potential blood hosts , both human and livestock , may also contribute to lower individual biting frequencies . The regional impact of climate change cannot be excluded , either , although the water delivery of the investigated rivers have not changed drastically until 1980 [26] . The longitudinal monitoring in the Vina du Nord valley indicates that the average transmission of O . volvulus remained around 500 L3 per man and year for 20 years after the onset of ivermectin mass treatments ( Fig 4A ) . This seems to contradict the reduction of onchocerciasis-positive patients in the region as a result of control strategies with ivermectin [12] . One reason could be that there is a variable degree of misidentification of O . volvulus L3 . In the most recent monitored years 2009–2012 , when molecular detection methods were used , the degree of morphological misidentification was 72% . However , earlier epidemiological data from the Sudan savannah [25] showed ATP above 4000 L3 per man and year ( Fig 4A , left side ) . Even though no differentiation of the species had been undertaken at that time , the proportion of animal-filariae in S . damnosum s . l . were likely low due to the lack of cattle as potential blood hosts [for explanation see 36] . Hence , only filariae from the warthog could have been co-transmitted , and the infection rate of the vector with O . ramachandrini has not changed considerably during the observation period . Another theory states that the regulation of parasite transmission may be density-dependent instead of linear . That means the effective reproductive ratio of filarial worms equals one even though the basic reproductive ratio is much higher . In the Vina du Nord river basin , Renz [25] compared the prevalence of onchocerciasis and burden of microfilariae with the L3 infection rate in the vector and found no linear correlation , but rather a dependency of fly infection rate with prevalence in the human population instead of the community's microfilarial load ( mff/mg ) . A density-dependent mechanism has already been shown for the parasite acquisition in cattle when inoculated with infective larvae of O . ochengi [37 , 38] . The observed seasonal variations of the entomological parameters match well with baseline data from the Vina du Nord river [25] , including the number of infective flies with L3 . Nonetheless , the O . volvulus ATP has drastically reduced to 3 . 5% of the baseline value meaning that the majority of infective flies now carry filariae of animal origin . Additionally , the number of L3 per infective fly decreased moderately ( from 3 . 2 to 1 . 8 ) . The low but stable transmission level of O . volvulus could mean that the threshold for maintaining endemicity is perhaps lower than current mathematical models predict ( ATP ≥ 100 in West Africa [39 , 40] , but also ATP ≥ 54 in Central America [41] ) . Even though molecular techniques of identification give higher accuracy , they are less suitable for high throughput analysis due to limitations of time , cost-effectiveness and local infrastructure . They are nonetheless very useful for the detection of unknown strains and species of filarial nematodes in vector and host , such as Onchocerca ochengi 'Siisa' [22 , 23 , 34] . Experimental infection studies from Togo [42] and Mali [43] , where O . ochengi microfilariae were inoculated by the vector from infected cattle , revealed shorter L3 body lengths ( Togo: 680 μm , 540–680; Mali: 647 μm , 540–810 ) than our observations ( 740 μm , 600–940 ) and previous studies from Cameroon [44] . These values , however , lie in proximity to the measurements for O . ochengi 'Siisa' ( 660 μm , 600–900 ) and may thus represent or morphologically resemble this strain . Ultimately the evolutionary relationships of Onchocerca parasites in humans , cattle and game animals can be compared and tested with genetic markers by generating phylogenetic trees [22 , 34] . Besides the climatic disparities of the two foci , which is mainly a result of different altitudes , they share similar conditions for their respective black fly populations . One major difference , however , is the disproportionately higher cattle stock density on the Adamawa plateau compared to the situation in the Sudan savannah ( Fig 1 ) . The cattle to human ratio around the Galim focus is approx . 10:1 , whereas in the Soramboum focus it ranges between 1:1 and 2:1 , and was even lower in previous years , because nomadic cattle were not allowed to enter the Vina du Nord basin until 1975 , and the local villagers had not kept any livestock animals , either . Nowadays , an increasing number of vagrant Bororo herdsmen arrive with their herds of zebu cattle and become settled . The inherent difference in livestock density is both culturally inherited ( migrating pastoralists of the North vs . settled cattle farmers of the South ) and due to biologic conditions ( water and food scarcity during the dry season in the Sudan savannah; absence of tsetse flies on the Adamawa plateau , which transmit bovine trypanosomiasis ) . Invading O . ochengi L3 elicit a humoral immune reaction in humans , which cross-reacts with O . volvulus L3 antigens , thereby reducing transmission success [19] . The protective effect of populations under repeated antigen exposure is called premunition and well known for malaria and other infectious diseases [45 , 46] . On the Adamawa plateau this effect seems to be strong enough to prevent or at least complicate the regional endemicity of O . volvulus . The advent of nomadic herdsmen and their cattle herds in the Vina du Nord valley is congruent with an increased transmission of animal-borne filariae , in particular O . ochengi ( Fig 4A ) . This sudden jump of animal-filariae in the vector population implies the diversion of large quantities of local S . damnosum s . l . to take their blood meal from cattle , thereby reducing the vector pool for humans [16] . This phenomenon has been termed zooprophylaxis and acts also as a protective trait against onchocerciasis transmission [15 , 20] . The important question is how the low but stable rate of onchocerciasis transmission in the Sudan savannah can be further curbed or completely stopped . Altogether , five molecularly identified O . volvulus L3 from two infective flies ( 3 . 85% of total , 95% CI: 0 . 47–13 . 21% ) were found in the dry season of 2010 ( February ) and the rainy season of 2011 ( June ) . Since yearly CDTI application rounds are given at the end of July , the late time point of occurrence after ivermectin treatment may hint to an incipient reconstitution of skin microfilariae in humans infected with O . volvulus 12 months prior . This would be a strong argument for the continuation or even temporary intensification of the ivermectin control program [10 , 36 , 47] . However , current political instabilities in adjacent countries and the exclusion of certain patient groups in treatment intervention programs , like nomadic people , illegal immigrants and refugees , could impede the long-term success of such measures . Ongoing monitoring of vector transmission is therefore crucial for health policy in onchocerciasis-endemic countries . | Over the past decades the Fight against river blindness , a tropical disease caused by a nematode worm , has been relatively successful , and a number of countries have been reported to be free of parasite transmission . In North Cameroon , we checked the occurrence of infective stages of Onchocerca volvulus in the transmitting black fly populations for more than three years and were able to confirm that the transmission there is low , but not yet interrupted . In a second location on a highland plateau , however , no infective stages of the human parasite were found . Instead , a closely-related parasite of cattle was present in both places . Given that the areas are not far away from each other and the biting frequencies of the black fly populations are similar , the historically earlier and higher density of cattle herds in one of the regions would explain why it is now free of the parasite due to the effects called zooprophylaxis and cross-reacting premunition . Changes in the socio-economic environment , especially the increase of human and cattle populations have a strong influence on the spread of river blindness in Africa . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [
"invertebrates",
"livestock",
"medicine",
"and",
"health",
"sciences",
"ecology",
"and",
"environmental",
"sciences",
"onchocerca",
"volvulus",
"rivers",
"ruminants",
"tropical",
"diseases",
"geographical",
"locations",
"vertebrates",
"vector-borne",
"diseases",
"parasitic",
"diseases",
"animals",
"mammals",
"onchocerca",
"developmental",
"biology",
"aquatic",
"environments",
"bodies",
"of",
"water",
"neglected",
"tropical",
"diseases",
"onchocerciasis",
"africa",
"infectious",
"diseases",
"cameroon",
"marine",
"and",
"aquatic",
"sciences",
"agriculture",
"people",
"and",
"places",
"helminth",
"infections",
"larvae",
"freshwater",
"environments",
"earth",
"sciences",
"nematoda",
"biology",
"and",
"life",
"sciences",
"metamorphosis",
"cattle",
"bovines",
"organisms"
] | 2016 | Ongoing Transmission of Onchocerca volvulus after 25 Years of Annual Ivermectin Mass Treatments in the Vina du Nord River Valley, in North Cameroon |
Campylobacter jejuni is the leading cause of human gastroenteritis worldwide with over 500 million cases annually . Chemotaxis and motility have been identified as important virulence factors associated with C . jejuni colonisation . Group A transducer-like proteins ( Tlps ) are responsible for sensing the external environment for bacterial movement to or away from a chemical gradient or stimulus . In this study , we have demonstrated Cj1564 ( Tlp3 ) to be a multi-ligand binding chemoreceptor and report direct evidence supporting the involvement of Cj1564 ( Tlp3 ) in the chemotaxis signalling pathway via small molecule arrays , surface plasmon and nuclear magnetic resonance ( SPR and NMR ) as well as chemotaxis assays of wild type and isogenic mutant strains . A modified nutrient depleted chemotaxis assay was further used to determine positive or negative chemotaxis with specific ligands . Here we demonstrate the ability of Cj1564 to interact with the chemoattractants isoleucine , purine , malic acid and fumaric acid and chemorepellents lysine , glucosamine , succinic acid , arginine and thiamine . An isogenic mutant of cj1564 was shown to have altered phenotypic characteristics of C . jejuni , including loss of curvature in bacterial cell shape , reduced chemotactic motility and an increase in both autoagglutination and biofilm formation . We demonstrate Cj1564 to have a role in invasion as in in vitro assays the tlp3 isogenic mutant has a reduced ability to adhere and invade a cultured epithelial cell line; interestingly however , colonisation ability of avian caeca appears to be unaltered . Additionally , protein-protein interaction studies revealed signal transduction initiation through the scaffolding proteins CheV and CheW in the chemotaxis sensory pathway . This is the first report characterising Cj1564 as a multi-ligand receptor for C . jejuni , we therefore , propose to name this receptor CcmL , Campylobacter chemoreceptor for multiple ligands . In conclusion , this study identifies a novel multifunctional role for the C . jejuni CcmL chemoreceptor and illustrates its involvement in the chemotaxis pathway and subsequent survival of this organism in the host .
Campylobacter jejuni is the one of the prevalent causes of acute human bacterial gastroenteritis worldwide [1]–[3] . C . jejuni is commonly found in the gastrointestinal tract of birds and poultry as commensal microbial flora [1] , with infections in humans usually occurring from consumption of undercooked poultry , unpasteurised milk or untreated water . Symptoms include development of abdominal pains , fever and diarrhoea which can contain blood and leukocytes [4] , [5] . Additionally Campylobacter enteritis is associated with post infectious complications ranging from reactive arthritis or reactive myositis to the more severe Guillain-Barré syndrome [1] , [6]–[8] . The development of disease depends on the ability of bacteria to adapt to the environment of the human gut [9] . To date , it is known that factors associated with virulence and pathogenicity of C . jejuni include iron acquisition , chemotaxis , adherence and lipooligosaccharides ( LOS ) [5] , [10] , [11] . C . jejuni is known to be highly motile in viscous environments , with motility and swimming velocity increasing with increasing viscosity [12] . Chemotaxis , the ability of bacterial cells to detect temporal changes in the chemical concentration of their surrounding environment , and flagella-mediated motility , have been reported to play an important role in the intestinal colonisation of avian and mammalian hosts , as well as the invasion of intestinal epithelial cells [10] , [12]–[17] . Furthermore , Hugdahl et al . ( 1988 ) have identified chemoattractants that C . jejuni is preferentially motile towards , which include amino acids found in the gastrointestinal tract , organic acid intermediates of the TCA cycle and components of mucous such as mucin . [18] . The importance of C . jejuni chemotaxis and motility in colonisation and pathogenicity has previously been shown with non-motile mutants defective in chemotaxis , unable to colonise and cause disease in the gastrointestinal tract of mice [13] , [19]–[21] . Furthermore , strains with mutations in chemotactic genes and flagella associated genes were not capable of colonising chicken caeca [14] and lost the ability to colonise rabbits [22] and ferrets [20] . Another important observation was that mutants defective in chemotactic motility also lost the ability to autoagglutinate , adhere and invade mammalian cells [12] , [15] , [16] , [23] , [24] . The fundamental components of the chemotaxis signalling pathway , which are conserved in all motile prokaryotes , consist of the chemoreceptors , a cytoplasmic histidine kinase , CheA , a coupling or scaffolding protein , CheW and/or CheV and a response regulator , CheY . In the well-characterised chemotaxis system of E . coli , CheA is regulated by the chemoreceptors through association with CheW and uses ATP to autophosphorylate a specific histidine residue . The phosphoryl group is subsequently transferred to the response regulator CheY [25] . Phospho-CheY interacts with the flagella motor to induce clockwise ( CW ) rotation resulting in a tumbling motion , where the cell momentarily stops and randomly reorientates [25]–[28] . Attractant-bound chemoreceptors inhibit CheA kinase activity , with a resulting decrease in the levels of phospho-CheY . In contrast , repellent-bound or empty chemoreceptors stimulate CheA kinase activity , thereby increasing phospho-CheY levels and tumbling events , orientating the cell in a new direction [25] , [26] , [28] , [29] . It is important to note that in C . jejuni , although the basic chemotaxis pathway backbone , consisting of Receptor-CheA-CheW ( V ) -CheY , is conserved , there are a number of differences to the paradigm E . coli model . C . jejuni encodes a two-domain CheA protein that includes a CheY-like response regulator domain in addition to the traditional histidine kinase domain , and a two domain CheV protein consisting of a CheW-like scaffolding domain and a CheY-like response regulator domain , as well as the paradigm CheW protein . C . jejuni also has a unique CheB protein that lacks a CheY-like response regulator domain found in all other bacterial chemotaxis pathways characterised to date [30] . Ten chemoreceptors have been identified in C . jejuni with homology to the methyl-accepting chemotaxis proteins ( MCPs ) in E . coli and have been designated Transducer-like proteins , Tlps . Group A Tlp receptors include CcaA ( Tlp1 ) , Tlps 2 , 3 , 4 , 7 and 10 which have similarities to E . coli MCP structures and family A transducers of Halobacterium salinarium [30] , [31] . Group A Tlps are thought to sense extracellular ligands and consist of a periplasmic sensory domain which is variable between different receptors , two transmembrane domains , and the highly conserved C-terminal cytoplasmic signalling domain [30] , [32] . The sensory domain of each Tlp appears to be unique from that of non-Epsilon proteobacterial chemoreceptors; therefore sequence homology alone is not sufficient to determine specific ligand or ligands for each of the receptors . To date two of the group A Tlp receptors , Tlp1 and Tlp7 , have been characterised [33] . CcaA was identified as the only receptor conserved in all sequenced strains of C . jejuni [34] and determined to be the receptor for aspartate [35] whereas Tlp7 binds to formic acid [36] . In this study we describe the characterisation of the periplasmic sensory domain of Cj1564 ( Tlp3 ) chemoreceptor of C . jejuni strain NCTC 11168-O and characterise the ligand binding specificities of this chemoreceptor protein . We demonstrate the multifunctional role of Cj1564 in recognition of chemoattractants as well as chemorepellents , its association with the scaffolding proteins CheV and CheW and demonstrate the role of Cj1564 in cell to cell adhesion ( autoagglutination ) and biofilm formation , cell shape and host colonisation .
To gain a greater understanding of the C . jejuni chemotactic pathway , characterisation of ligand binding potential for Cj1564 ( Tlp3 ) was performed using recombinant Tlp3 periplasmic sensory domain peptide with all 20 amino acids and salts of organic acid arrays as well as 96-well binding assays ( Table 1 ) . The array assay presents the compounds covalently bound to an epoxide group resulting in a high probability of a single molecular orientation . The plate assay relies on non-covalent charge and hydrophobic/hydrophilic interactions for presentation of the molecules offering a different presentation to the array . A total of 12 interacting ligands were identified through the plate and array assays . The binding of the Tlp3 sensory domain to amino acids and salts was then confirmed by STD-NMR and SPR ( Biacore , Table 1 ) . SPR analysis found the highest affinity interactions of Tlp3 with lysine and glucosamine ( KD<10 µM; Table 1 ) , additionally , biologically significant interactions ( KD<50 µM; Table 1 , Figure S1 ) were observed for isoleucine , succinic acid , arginine , purine , malic acid and thiamine . Cj1564 ( Tlp3 ) has previously been identified as a group A chemoreceptor for C . jejuni with homology in the signalling domain to other bacterial chemoreceptors such as that of E . coli and H . pylori [37] . The periplasmic domain of C . jejuni group A chemoreceptors are predicted to be involved in sensing and binding of extracellular ligands [37] . In order to identify biologically significant ligand binding specificity and function of Tlp3 in the chemotactic pathway , an insertionally inactivated isogenic tlp3 mutant and its complement were created . The mutant was constructed by deleting 52 central bp of the periplasmic domain of tlp3 and inserting a non-polar kanamycin resistance cassette which has the transcription terminator removed to minimise effects on genes downstream that may be transcribed in the same orientation as tlp3 ( described in materials and methods: mutagenesis and complementation of tlp3 ) . Microscopic analysis of Δtlp3 confirmed the presence of flagella and altered ( loss of spiral shape ) cellular morphology of all observed 11168-O Tlp3 mutant bacterial cells , with spiral morphology restored after complementation in approximately 50% of the complemented cells ( Figure 1A , B & C ) . Comparison of C . jejuni 11168–O and Δtlp3 motility shows that there was a 5-fold decrease in non-directed swimming motility of the mutant with motility partially restored to wild type levels in the complemented strain , Δtlp3c ( Figure 1D ) . Live imaging of fluorescently labelled C . jejuni isogenic strains allowed capture of bacterial cell motility and demonstrated a defect in motility illustrated by the inability of Δtlp3 cells to effectively ‘swim’ in a directional movement in absence of specific stimuli . Instead the cells were observed to ‘twitch’ and continuously ‘tumble’ without movement in any particular direction ( Movie S1 ) , thus demonstrating an altered “random walk” motility phenotype of the Δtlp3 mutant . The expression of the tlp3 gene in the isogenic mutant strain was reduced , as compared with the wild type and complemented strains , which is likely to be due to the interruption of the gene with the strongly promoted KmR cassette . Expression of tlp3 was analysed using quantitative real time PCR in 11168-O , Δtlp3 and Δtlp3c using primers that amplify the entire periplasmic domain gene region of tlp3 . There was a 6 . 5±0 . 78 fold reduction observed in the expression of tlp3 in the Δtlp3 mutant strain compared to the wild type strain , conversely an 8 . 75±0 . 98 fold increase was observed when screening for tlp3 expression in the complemented mutant compared to wild type ( data not shown ) . This indicates that the inclusion of the resistance cassette into the tlp3 gene has negatively affected the expression of tlp3 from its own promoter . This indicated that the Kanamycin resistance cassette encodes the dominant promoter signal that inhibits production of RNA molecules containing the 5′ end of the tlp3 gene region , upstream from the kanamycin resistance cassette . While the insertion of complementing tlp3 and chloramphenicol resistance cassette into the pseudogene ( cj0046 ) results in over expression of tlp3 compared to the wild type strain . Further expression analysis was performed on the genes immediately upstream ( cj1563c ) and downstream ( pflA ) of the mutated tlp3 gene to confirm no polar effects were introduced as a result of mutagenesis . Expression of pgp1 and flaA were assessed to establish that the observed defect in motility is not due to changes in expression of genes involved in flagella development or function . No significant difference in expression ( p>0 . 1 ) was observed for cj1563c , pflA , pgp1 and flaA ( data not shown ) . Biofilm formation and agglutination play an important role in the survival of bacterial cells . Consequently , we compared these characteristics between the wild type ( 11168-O ) , mutant ( Δtlp3 ) and complemented ( Δtlp3c ) strains . Autoagglutination assays revealed that the Δtlp3 autoagglutinated while 11168-O and Δtlp3c had little to no autoagglutination ( Figure 2A & B ) . It is also interesting to note that autoagglutination of the 11168-O Δtlp3 mutant was similar to that of C . jejuni 81–176 ( Figure S2 ) , which carries a natural mutation in its tlp3 gene , revealed by the published genome sequence ( Sanger , 2006 ) . Furthermore , autoagglutination was also shown to be independent of growth temperature at 25°C , 37°C and 42°C for all strains tested ( data not shown ) . Additionally it appears that Δtlp3 autoagglutinates at a rate that is faster than normal gravitational pull . Biofilm formation , as assessed by crystal violet assay , indicated an approximate 1 . 5-fold increase in biofilm formation of Δtlp3 mutant compared to wild type , with the original levels in biofilm formation restored in Δtlp3c ( Figure 2C ) . In order to determine the biological relevance of Tlp3 in chemotactic motility , nutrient depleted chemotaxis assays were performed as previously described [35] . Chemotaxis towards a range of amino acids , glycans and other small molecules was investigated in 11168-O , Δtlp3 and the complemented mutant , Δtlp3c . A positive chemotaxis response was identified for 5 amino acids , suggesting these ligands are attractants ( Figure 3A; Table 1 ) . A 4-log reduction in bacterial numbers was observed for the Δtlp3 ( 3 . 6 × 102 cfu/ml ) compared to the wild type ( 6 × 106 cfu/ml ) in migration towards isoleucine , additionally this was also observed for fumaric acid ( 9 × 102 and 4 . 14 × 106 cfu/ml respectively ) . For migration towards purine , a 1-log reduction for the Δtlp3 ( 1 . 08 × 105 cfu/ml ) compared to the wild type ( 1 . 4 × 106 cfu/ml ) was observed , with a 2-log reduction in migration towards malic acid ( 1 . 13 × 103 and 9 × 105 cfu/ml respectively ) and a 3-log reduction in migration towards aspartate ( 3 . 7 × 103 and 7 . 8 × 106 cfu/ml respectively ) . The migration of wild type C . jejuni 11168-O towards aspartate was comparable to that previously published for nutrient depleted assay by Hartley-Tassell et al . , 2010 [38] . Tlp3 was also found to mediate a repellent response to 5 amino acids ( Figure 3B; Table 1 ) . Lysine mediated repellence was reduced for the Δtlp3 isogenic strain as a 2-log increase in viable bacterial numbers was detected around the lysine plug ( 2 . 4 × 105 cfu/ml ) when compared to the wild type ( 2 . 2 × 103 cfu/ml ) . For migration of Δtlp3 towards glucosamine , a 3-log increase ( 7 . 8 × 106 cfu/ml ) was detected , compared to that of the wild type ( 1 . 68 × 103 cfu/ml ) . In addition , for succinic acid , arginine and thiamine , a 1-log increase of bacterial numbers was observed in the Δtlp3 ( 2 × 107 , 6 . 8 × 107 and 9 × 106 cfu/ml , respectively ) compared to the wild type strain ( 1 . 5 × 106 , 1 . 3 × 106 and 4 × 105 cfu/ml , respectively ) . Chemotaxis assay for alpha-ketoglutarate was not definitive ( data not shown ) , possibly due to the low affinity of this receptor for alpha-ketoglutarate . 1H STD-NMR analysis was used to investigate the epitope binding preferences of the chemoattractants ( isoleucine , purine , malic acid ) and chemorepellents ( lysine , arginine , glucosamine ) to the recombinant Cj1564 ( Tlp3 ) periplasmic sensory domain peptide . Binding was observed with chemoattractants isoleucine and purine along with the chemorepellents lysine and arginine in accordance with the SPR analysis showing varying binding affinities . For the chemoattractant isoleucine ( KD∼17 µM ) , the binding epitope appeared to be the methyl group , as strong methyl resonances were seen in the STD spectrum and only very weak signals for other side chain resonances ( Figure 4A ) . For purine ( KD∼38 µM ) , the H6 and to a lesser extent the H2 protons ( ∼5-fold ) of the pyrimidine ring showed an STD effect compared to a negligible STD effect for the H8 proton of the imidazole ring . For the chemorepellent lysine ( KD∼2 . 8 µM ) an STD effect was clearly seen for all proton resonances along the side chain ( Figure 4B ) . A weaker STD effect ( ∼4 fold ) was seen for all proton resonances along the arginine ( KD∼38 µM ) side chain . No signals were detected for malic acid or glucosamine using STD-NMR . SPR analysis demonstrated that glucosamine and malic acid have slow disassociation rates explaining the absence of STD signals due to these amino acids as saturation transfer is a result of the spin diffusion process which requires fast exchange for magnetisation to be spread from the protein to the ligand in order to observe an STD effect . Competition STD experiments were also performed to probe the binding preferences of Tlp3 and to unravel the nature of the binding interaction . Competition experiments testing the effect on binding to Tlp3 of attractant/attractant ( isoleucine/purine ) , repellent/repellent ( lysine/arginine ) and attractant/repellent ( isoleucine/arginine; isoleucine/lysine; purine/lysine; purine/arginine ) were performed . For attractant/attractant ( isoleucine KD∼17 µM/purine KD∼38 µM ) there seemed to be a minor change in the STD effect seen for the ligands independently . For the repellent/repellent ( lysine KD∼2 . 8 µM/arginine KD∼38 µM ) however , there appeared to be a significant reduction in the STD effect of lysine in the presence of arginine whose weak STD effect appeared unchanged , even though the binding affinity of lysine is stronger ( ∼14 fold ) than that of arginine . These results demonstrate that Cj1564 ( Tlp3 ) is able to bind to both chemoattractants and chemorepellents . No binding preference to Cj1564 ( Tlp3 ) was observed in the presence of the chemoattractants isoleucine/purine , compared with the ligands alone , however for the chemorepellent competition experiment even though lysine has a greater binding affinity ( KD∼2 . 8 µM ) compared with isoleucine ( KD∼17 µM ) a significant reduction in the STD effect of lysine was observed in the presence of arginine suggesting preferential binding of arginine over lysine . In the case of the attractant/repellent spectra: Isoleucine/arginine - isoleucine ( chemoattractant , KD∼17 µM ) binds preferentially over arginine ( chemorepellent , KD∼38 µM ) . For isoleucine/lysine - isoleucine ( chemoattractant ) binds preferentially over lysine ( chemorepellent ) with a significant reduction in the STD effect of lysine observed in the presence of isoleucine compared to the STD spectra of the ligands individually , even though the binding affinity of lysine ( KD∼2 . 8 µM ) is stronger than that of isoleucine ( KD∼17 µM ) ( Figure 4C & D ) . Purine/lysine - purine ( chemoattractant ) binds preferentially over lysine ( chemorepellent ) with a significant reduction in the STD effect of lysine in the presence of purine observed compared to the STD spectra of the ligands individually , even though the binding affinity of lysine ( KD∼2 . 8 µM ) is stronger than that of purine ( KD∼38 µM ) . Purine/arginine - purine ( chemoattractant ) binds preferentially over arginine ( chemorepellent ) , a negligible STD effect of arginine in the presence of purine was observed with these ligands determined to have comparable binding affinities ( KD∼38 µM ) . These results demonstrate that there is preferential binding of chemoattractants over the chemorepellents regardless of the individual binding affinity of the ligands to Cj1564 ( Tlp3 ) . To further investigate the role of Tlp3 in the chemotaxis signal transduction pathway in C . jejuni , the predicted cytoplasmic signalling domain of Tlp3 was analysed for protein-protein interactions with C . jejuni NCTC11168-O chemotaxis proteins using the yeast two-hybrid and three-hybrid systems as previously described [35] . The yeast two-hybrid system was used because of a lack of similar genetic manipulation systems available for C . jejuni . We have previously demonstrated biological validity of protein interactions identified by the yeast two-hybrid system via pull-down assays for the components of the C . jejuni aspartate receptor , CcaA [35] signal transduction . The yeast system allows detection of interacting proteins in vivo by utilising the two separable domains of the GAL-4 transcription factor , the DNA-binding domain ( DNA-BD ) and the transcription activation domain ( AD ) . This system relies on the reconstitution of the GAL-4 transcription factor when two proteins of interest , the ‘bait’ protein fused to the DNA-BD and the ‘target’ protein fused to the AD , interact , thus allowing activation of reporter gene expression [40] , Residues 517–662 of Tlp3 , encompassing the region homologous to the highly conserved MCP signalling domain , were selected for analysis . The signalling domain is an independent structural motif , which interacts with the dimeric CheA and the scaffolding protein CheW to form MCP-CheW-CheA ternary signalling complexes that regulate CheA histidine kinase activity [41]–[43] . It is necessary to note that this region of Tlp3 is identical to residues 513–659 of Tlp2 ( Cj0144 ) and 520–665 of Tlp4 ( Cj0262 ) [37] and thus is referred to as Tlpsig2 . 3 . 4 in this study ( Figure S5 ) . The Tlpsig2 . 3 . 4 was found to interact with a number of the C . jejuni NCTC11168-O chemotaxis proteins and protein domains , as shown in Table 2 . A medium strength interaction was detected between Tlpsig2 . 3 . 4 and CheV ( AD-Tlpsig2 . 3 . 4 with BD-CheV ) . An identical interaction was observed in the reciprocal comparison of fusion proteins ( AD-CheV and BD-Tlpsig2 . 3 . 4 ) . The Tlpsig2 . 3 . 4 was found to bind only to the CheW-like domain of CheV ( CheVdW ) . A medium strength interaction was detected between these domains in reciprocal comparisons ( AD-Tlpsig2 . 3 . 4 with BD-CheVdW and AD-CheVdW and BD-Tlpsig2 . 3 . 4 ) while no interaction was detected with the response regulator domain of CheV ( CheVdRR and Tlpsig2 . 3 . 4 ) . Two-hybrid analysis also revealed that the Tlpsig2 . 3 . 4 signalling domain was capable of dimerisation ( AD-Tlpsig2 . 3 . 4 and BD-Tlpsig2 . 3 . 4 ) . An interaction was also observed between Tlpsig2 . 3 . 4 and CheW . This interaction was of a medium strength ( AD-CheW and BD-Tlpsig2 . 3 . 4 ) ; however this interaction was not detected with the reciprocal combination of fusion proteins . As it has previously been shown that CcaA ( Tlp1 ) of C . jejuni preferentially binds to CheV over CheW [35] , we utilised the yeast three-hybrid system to determine if the signalling domain of Tlp3 ( which is identical to the signalling domains of Tlp2 and Tlp4 ) displayed similar binding preferences for the scaffolding proteins . In the three-hybrid system , a third protein can be conditionally expressed and its role in the interaction between the AD and DNA-BD fusion proteins can be examined [44] . We specifically looked at the effect of presence of both CheW and CheV proteins on their interactions with Tlpsig2 . 3 . 4 . It was found that the co-expression of native CheW from pBrTlp234IWII ( Table 3 ) had no effect on the medium strength interaction observed between Tlpsig2 . 3 . 4 and CheV ( BD-Tlpsig2 . 3 . 4 and AD-CheV ) . Analysis of interactions between the reciprocal combinations of fusion proteins using pBrVIWII ( Table 3 ) showed that the co-expression of native CheW slightly enhanced the medium strength interaction of CheV with Tlpsig2 . 3 . 4 ( BD-CheV and AD-Tlpsig2 . 3 . 4 ) . When protein interactions of Tlpsig2 . 3 . 4 with CheW were investigated with native CheV co-expressed ( pBrTlp234IVII ) , it was found that the presence of CheV strengthened the weak interaction observed between the Tlpsig2 . 3 . 4 signalling domain and CheW ( BD-Tlpsig2 . 3 . 4 and AD-CheW ) . This was confirmed in the analysis of the interactions between the reciprocal combination of fusion proteins ( AD-Tlpsig2 . 3 . 4 and BD-CheW ) using pBrWIVII .
In this study we have shown the role of Tlp3 in C . jejuni 11168-O chemotaxis as a multiple ligand-binding protein , capable of detecting numerous chemoattractants and chemorepellents using a range of confirmatory methodologies such as small molecule arrays , SPR and STD-NMR . The substrates identified as having a strong binding affinity include lysine and glucosamine with biologically significant interactions observed for isoleucine , aspartate , succinic acid , arginine , purine , malic acid , and thiamine . Therefore we propose to name this receptor CcmL; Campylobacter chemoreceptor for multiple ligands . Similar observations have previously been made for H . pylori isolates , where positive chemotaxis was identified for the sugars and amino acids phenylalanine , aspartic acid , glutamic acid , isoleucine and a negative chemotaxis for leucine and tyrosine [45] . Isoleucine has previously been demonstrated to be a chemoattractant for other organisms including B . subtilis [46] , H . salinarum [47] , P . aeruginosa [48] and H . pylori [45] . In B . subtilis , the McpC chemoreceptor binds isoleucine weakly , yet with sufficient affinity to suggest direct binding . It is interesting to note that in our study , glucosamine has been identified as a repellent , while in other bacteria such as E . coli [49] and B . burgdorferi it has been determined to be an attractant , and has been classified as a non-essential nutrient [48] , [50] . Lysine has been demonstrated as an attractant for P . aeruginosa [48] which is opposite to our findings which show lysine as a chemorepellent for C . jejuni . Furthermore , in B . subtilis , the McpC chemoreceptor did not show binding to lysine , however , it was suggested that McpC binds lysine by an indirect method most likely involving ancillary proteins , further suggesting McpC may be a universal chemoreceptor able to respond to numerous amino acids [51] . It had been shown in E . coli that even though chemoreceptors are sensitive to a particular ligand , they can also detect a large number of structurally related amino acids and their analogues [52] . C . jejuni is highly adapted to the environment of the avian gut and as a result , uses efficient chemotactic motility to colonise the mucous-filled crypts of the lower gastrointestinal tract [53] . It is reasonable to hypothesise that Tlp3 may be involved in interacting with ligands to sense the external environment in order to navigate inside the host and may be involved directly or indirectly in C . jejuni metabolic and catabolic pathways . It is interesting to note that aspartate , glutamate , proline and serine are the most abundant amino acids found in chicken excreta [54] , and serine catabolism has been reported to be essential for colonisation of the avian gut by C . jejuni [55] . However , C . jejuni lacks the key glycolytic enzyme 6-phosphofructokinase as well as alternative pathways for sugar catabolism [56] so it utilises amino acids and Krebs cycle intermediates for energy production and encodes all of the enzymes required for a complete oxidative TCA cycle [37] , [57] , [58] . C . jejuni needs to be able to sense and move towards amino acids and small organic acids such as aspartate , asparagine and serine in order to catabolise these compounds to use as the sole source of reduced carbon and energy due to its inability to utilise glucose [59] . Additionally , a recent study has indicated that energy taxis may also be one of the driving forces behind movement to optimal conditions for energy generation and subsequent colonisation [60] , [61] . We have shown that a mutation in the C . jejuni ccmL chemoreceptor gene lead to alteration of phenotypic characteristics of the bacteria , such as cellular morphology , autoagglutination behaviour and biofilm formation , highlighting its role in the C . jejuni life cycle . Furthermore , the signalling domain of this chemoreceptor also interacts and binds with both CheV and CheW scaffolding proteins . Additionally , we used STD-NMR analysis to confirm the binding of amino acid and salts of organic acid arrays , plate binding assays and SPR analysis . STD-NMR indicated that the Tlp3 receptor binds chemoattractants preferentially over chemorepellents irrespective of binding affinity , however , is still able to recognise both attractants and repellents in isolation . This may indicate that upon binding an attractant , the repellent binding site of CcmL through allostery prevents the binding of a repellent ligand to the protein . Furthermore , it appears that the binding site may be able to accommodate more than one chemoattractant at a time as competition tests between two chemoattractants results in positive binding for both attractants present; indicating that one chemoattractant is not preferred over another regardless of the measured affinity . These results indicate a much more complex interaction between chemotaxis receptor proteins and ligands than previously thought , with a complex milieu of attractants acting in concert rather than the receptor binding preferentially to a specific ligand based on the hierarchy of affinities . Furthermore , the Tlp3 predicted structure indicates the presence of a single Cache_1 ( PFO2743 ) ( calcium channels and chemotaxis receptors ) domain , representing a single ligand binding pocket that accommodates multiple ligands with varying affinity . This is in agreement with our STD-NMR data that shows interchange of the ligands binding to purified periplasmic domain of CcmL in competitive assays . CcmL ( Cj1564 ) shares complete homology in the cytoplasmic domain with Tlp2 ( Cj0144 ) and Tlp4 ( Cj0262 ) , but has less than 50% homology with the cytoplasmic domains of Tlp1 ( Cj1506 ) , Tlp7 ( Cj0951/52 ) and Tlp10 ( Cj0019 ) . Tlp3 has the greatest homology to Tlp2 with 72% identity across the entire protein and 38% identify/59% similarity across the periplasmic domain . The greatest stretch of similarity between Tlp3 and Tlp2 in the periplasmic binding region is within the cache domain that is present in both proteins . Homology between the periplasmic domain of Tlp3 and the periplasmic domains of Tlp1 and 4 is below 33% . No homology was detected between Tlp3 periplasmic domain and the periplasmic domains of Tlp7 and 10 . To further characterise interactions of Tlp3 with components of the C . jejuni chemotaxis signalling pathway , a well-established yeast-two-hybrid and three-hybrid protein-protein interaction system was used . The two- and three-hybrid system data suggests that both the scaffolding proteins CheW and CheV are capable of binding to the Tlps2 , 3 and 4 signalling domains with no obvious preference for either protein observed . CheW and CheV may form mixed multi-protein complexes with these receptors , as the expression of CheV was found to strengthen the interaction between the signalling domain of Tlpsig2 . 3 . 4 and CheW . This appears to vary from the observations for the CcaA ( Tlp1 ) signalling domain , which has previously been shown to preferentially bind CheV [35] . While the signalling domains of Tlps 2 , 3 and 4 and of Tlp1 are very similar , some amino acid differences do exist which may account for the differences in the binding preferences of these Tlps for the CheW and CheV scaffolding proteins . These amino acid differences are within the region homologous to residues 350–471 of the E . coli serine chemoreceptor , Tsr , a fragment which has been shown to be capable of mediating CW-biased signals and is therefore predicted to contain residues involved in the binding of CheW and CheA in this species [62] ( Figure S5 ) . This data lead us to speculate that in C . jejuni , the group A Tlp2 , 3 and 4 signal via alternative or mixed scaffolding proteins and that different binding affinities of the chemoreceptors with CheV and CheW may control the composition of receptor clusters . The mutation in the tlp3 ( cj1564 ) gene resulted in an altered cellular morphology of the bacterial cell and inability of Δtlp3 cells to ‘run’ . Increased formation of biofilm and the high rate of autoagglutination may indicate a possible increased response to stress by the tlp3 mutant . A previous study by Vegge et al . ( 2009 ) reported that mutation of the tlp3 gene along with other genes does not affect the motility of C . jejuni in a rich medium [63] however; our results indicate mutation of tlp3 significantly reduces motility when compared to the wild type and complemented strains . This is in agreement with a study carried out by Golden et al . ( 2002 ) [64] where mutation of tlp3 reduced motility and complementation restored the wild type phenotype [23] . Furthermore , reduced motility is also likely due to an inability of the bacteria to engage in efficient chemotaxis signalling . Further inspection of the isogenic Δtlp3 mutant cells by microscopy indicated that the bacteria retained the ability to twitch and tumble in a stationary position suggesting a bias toward clockwise rotation of the flagella , subsequently leading to a higher frequency of tumbles and inhibited smooth swimming . It appears however , that the bacterial cell was still capable of directional movement when a stimulus was supplied in vivo . Additionally , scanning electron microscopy and quantitative PCR analysis confirmed the presence of flagella indicating that the lack of motility was not due to a defect in the flagella development; however , the helical shape of the bacteria was altered . The helical shape of C . jejuni has long been associated with pathogenesis but the genetic components involved in modulating C . jejuni morphology have only recently been identified , where the protein termed Pgp1 ( peptidoglycan peptidase 1 ) in C . jejuni 81–176 was identified to be involved in maintenance of the helical shape [65] . To date no correlation has been found between chemoreceptors and cellular morphology . In the in vitro model of infection , it appears that Tlp3 may be a critical factor for invasion as Δtlp3 displayed a markedly reduced level of adherence and invasion compared to the wild type strain . However , in vivo assessment of the Δtlp3 mutant and wild type strain showed no differences in colonisation ability of the avian caeca based on cell counts , even though ability to adhere and invade caco-2 cells in vitro was significantly reduced . This finding agrees with the previously published data on avian colonisation for the Tlp3 mutant of C . jejuni 81–176 [66] . However , it is important to consider that subsequent publication of the genome sequence of 81–176 ( Sanger , 2006 ) revealed a natural mutation of the Tlp3 homologue in 81–176 which exists as 2 separate reading frames: CJJ81176_1548 and CJJ81176_1549 encoding the majority of the periplasmic and cytoplasmic domains of Tlp3 , respectively . There is a nucleotide deletion in the sequence of CJJ81176_1548 at position 1467624 ( A ) in the genome subsequently altering the amino acid sequence , consequently changing the reading frame of the transmembrane and cytoplasmic domains ( CJJ81176_1549 ) resulting in incorrect translation and production of a non-functional Tlp3 protein in 81–176 . There are an additional number of mutations in both the remainder of the periplasmic and cytoplasmic domains with several intervening stop codons . The absence of a functional Tlp3 in 81–176 has previously been reported [67] . It is interesting to note that C . jejuni 81–176 has the same complement of Tlp genes as 11168 and 81116 , with the exception of Tlp3 , which has a naturally occurring mutation in 81–176 . When phenotypic characteristics pertaining to biofilm formation and autoagglutination of 81–176 , and 11168-O are compared , a wild type 81–176 shows higher levels of aggregation , biofilm formation ( data not shown ) and autoagglutination , similar to that observed for Tlp3 mutant of 11168 ( Figure S2 ) . Characterising the function of C . jejuni chemosensory proteins , as described in this study , will contribute to understanding chemotaxis signalling pathways which are involved in colonisation and further identify chemoreceptor ligand specificity of individual group A Tlp receptors and their involvement in the chemotaxis pathway and its importance in the survival of this organism . The findings in this study also provide insight into the complexity of chemotaxis receptor protein-ligand interactions with implications not just for C . jejuni chemotaxis but for all bacterial chemotaxis .
Animal experiments were carried out in strict accordance with the Griffith University Animal Ethics Committee guidelines and assigned approval number BDD/02/11 . All procedures involving animals were reviewed and approved by National Health and Medical Research Council Australian code of practice for the care and use of animals for scientific purposes 7th edition 2004 . Bacterial strains , yeast strains and plasmids used in this study are described in Table S1 . C . jejuni strain 11168-O from Skirrow collection was kindly provided by DG Newell , Central Veterinary Laboratories , UK . C . jejuni strains were grown at 37°C or 42°C on Columbia agar supplemented with 5% defibrinated horse blood ( HBA ) with vancomycin ( 10 µg/ml ) , trimethoprim ( 2 . 5 µg/ml ) and polymyxin B ( 2 . 5 IU/ml ) under microaerobic conditions ( 5% O2 , 10% CO2 , 85% N2 ) for 18–24 h . Strains with plasmids for mutation or complementation studies were grown with 50 µg/ml kanamycin ( Km ) or 30 µg/ml chloramphenicol ( Cm ) . Host E . coli BL21 DE3 ( Novagen , USA ) and E . coli DH5α ( Novagen , USA ) strains were grown in Luria-Bertani ( LB ) medium ( Oxoid ) supplemented with ampicillin ( 100 µg/ml ) , kanamycin ( 50 µg/ml ) and chloramphenicol ( 30 µg/ml ) when required . Host yeast strains were grown and prepared according to manufacturers' instructions ( Clontech ) . Labelling of bacterial cells with CFDA-SE was performed as previously described [35] , [68] . The DNA fragment encoding the periplasmic sensory domain ( amino acids 43–290 ) of Tlp3 was amplified using primers incorporating start and stop codons ( Table S2 ) and ligated into pGEM-TEasy to form pGU0815 ( Table S1 ) . The insert in pGU0815 was restricted at primer-specific NdeI and XhoI sites and subcloned into pET-19b to form pGU0816 ( Table S1 ) . For recombinant protein expression , competent E . coli BL21 ( DE3 ) cells were transformed with pGU0816 . An overnight culture of BL21 ( DE3 ) /pGU0816 was used to inoculate LB containing ampicillin ( 100 µg/ml ) that was incubated at 37°C with aeration . Protein expression was induced using 1 mM IPTG when OD600 nm reached 0 . 4–0 . 6 . Expression of the Tlp3peri -His fusion protein was verified by SDS-PAGE and Western Blot analysis using Anti-His mouse IgG ( Cell Signalling ) as shown in Figure S6 . The cell pellet was resuspended in PBS containing 6 M urea , lysozyme ( 0 . 2 mg/ml ) and protease inhibitor cocktail mix ( 50 µl/ml ) and incubated at room temperature for 1 h using a rotational mixer . The cells were sonicated and an additional freeze/thaw step performed to aid in cell lysis . The insoluble cell debris was removed by centrifugation at 38 000 rpm for 90 min . The clarified supernatant was added to 1 . 5 ml of TALON Metal Affinity Resin ( Clontech ) and rotated overnight at 4°C using a rotational mixer . The slurry mix was then packed by gravity into a 10 ml Bio-Rad chromatography column . The column was washed twice with PBS , then washed with PBS containing 20 mM imidazole , washed three times with PBS and the bound His-tagged protein was eluted with PBS containing 150 mM imidazole . Residual imidazole was removed from the sample using Econo-Pac 10DG Desalting Columns ( Bio-Rad ) according to manufacturer's specification . Purity was confirmed by analysis of samples by SDS-PAGE and Western Blot using anti-His antibodies ( Bio-Rad ) . To ensure that the recombinant protein was properly folded , its properties were compared to denatured protein . Denatured protein recombinant Tlp3 was not soluble in PBS or TBS buffers , unlike correctly folded protein . Moreover , the binding affinities of recombinant protein following purification without urea were also tested . No change in binding affinities could be observed ( data not shown ) . Amino acid and small molecule arrays and plate based binding assays were performed as previously described . [35] . The ligands investigated by Surface Plasmon Resonance ( SPR ) included alpha-ketoglutarate , glucosamine , lysine , purine ( basic ring ) , malic acid and L-isoleucine ( Sigma ) . SPR experiments were performed using a Biacore T100 biosensor system ( GE- healthcare ) at 25°C in 1× PBS pH 7 . 2 at a flow rate of 30 µl/min . Purified His-Tlp3 was diluted to 0 . 15 µM in 1× PBS pH 7 . 2 and loaded on flow cell 2 ( FC2 ) of a Ni2+ NTA sensor chip with 5 min contact time . FC1 had no protein loaded and was used as reference . Amino acids were prepared in 1 × PBS pH 7 . 2 and serially diluted from 0 . 0125 to 0 . 2 mM . The amino acids were loaded to the sensor chip using single-cycle kinetics , i . e . after the injection of five amino acid dilutions the chip was regenerated with EDTA . Subsequently , the chip was re-loaded with Ni2+ and His-Tlp3 before the injection of the next amino acid to be tested . A 10 min dissociation time was allowed after the addition of each analyte . SPR signals were analyzed using the Biacore Evaluation software to determine KD . Recombinant Tlp3 ( 1 mg/mL , 100 µL ) dissolved in D2O ( 99 . 99% D Cambridge Isotopes ) containing 50 mM NaCl and 50 mM KH2PO4 and added to various ligands ( 1 mg/mL , 150 µL ∼150 mole equivalents ) also dissolved in D2O ( 99 . 99% D Cambridge Isotopes ) containing 50 mM NaCl and 50 mM KH2PO4 to give a total volume of 250 µL in a 3 mm NMR tube for NMR analysis . Control samples were prepared in an identical manner without Tlp3 added . All NMR experiments were performed on a Bruker Avance 600 MHz spectrometer , equipped with a 5-mm TXI probe with triple axis gradients at 283 K without sample spinning . 1H NMR spectra were acquired with 32 scans , a 2 s relaxation delay over a spectral width of 6000 Hz . Solvent suppression of the residual HDO peak was achieved by continuous low-power presaturation pulse during the relaxation delay . In the STD-NMR experiments of the chemoattractants ( isoleucine , purine and malic acid ) and chemorepellents ( lysine , arginine and glucosamine ) in complex with Tlp3 , the protein was saturated at −0 . 5 ppm in the aliphatic region of the spectrum and off-resonance at 33 ppm with a cascade of 40 selective Gaussian-shaped pulses of 50 ms duration ( 50 dB ) , which correlates to a strength of 190 Hz . A 100 µs delay between each soft pulse was applied , resulting in a total saturation time of 2 s and 2 K scans . Data were obtained with an interspersed acquisition of pseudo-two-dimensional on-resonance and off-resonance spectra in order to minimize the effects of temperature and magnet instability . On- and off-resonance spectra were processed separately , and the final STD-NMR spectrum was obtained by subtracting the individual on- and off-resonance spectra , resulting in less subtraction artefacts . Relative STD effects were calculated according to the equation ASTD = ( I0−Isat ) /I0 = ISTD/I0 by comparing the intensity of the signals in the STD-NMR spectrum ( ISTD ) with signal intensities of a reference spectrum ( I0 ) . The STD signal with the highest intensity was set to 100% and other STD signals were calculated accordingly . A spin lock field of 10 ms was applied to remove unwanted background protein signals . Increased spin lock fields resulted in artefacts and reduced ligand signal intensities . Control STD-NMR experiments were performed using an identical experimental setup and the same ligand concentration but in the absence of the protein . The tlp3 gene ( cj1564 ) in C . jejuni 11168-O was inactivated by inverse PCR mutagenesis [69] to produce the strain 11168-OΔtlp3::Km ( Table S1 ) . The tlp3 periplasmic domain ( amino acids 43–290 ) was amplified from the 11168-O genome and cloned into pGEM-TEasy to produce pGU0815 ( Table S1 ) . Inverse PCR with primers designed to incorporate a BglII restriction site and delete 52 bp of the tlp3 periplasmic domain was followed by insertion of a non-polar kanamycin resistance cassette with a consensus campylobacter promoter from pMW10 [70] in the same orientation as tlp3 to generate plasmid pGU0817 ( Table S1 ) . A kanamycin resistance cassette which lacks the transcription terminator was used in order to minimise effects on genes downstream of tlp3 . Each construct was verified by DNA sequencing and subsequently electro-transformed into the motile variant of 11168 original clinical isolate , donated by D . G Newell , VIR , London [71] . Replacement of the mutant allele was verified by PCR and DNA sequencing . The complement of this mutant was generated by inserting the complete tlp3 gene including promoter into the pseudogene cj0046 using the plasmid pC46 ( Table S1 ) . The tlp3 complement plasmid was electro-transformed into C . jejuni 11168-O with insertion into the pseudogene cj0046 locus to produce strain 11168-OΔtlp3::KmΩcj0046::Cm ( Table S1 ) . Complementation was confirmed by PCR and sequence analysis . Motility assays were performed as previously described [23] with the following modifications: C . jejuni strains were grown microaerobically at 42°C for 24 h in 10 ml Brucella broth ( Oxoid ) on a shaker ( 50 rpm/min ) . Cells were collected by centrifugation at 3000 g for 5 min and washed once with Brucella broth . Equal numbers of bacterial cells ( 5 µl ) of 1 × 109 cfu/ml of each strain of C . jejuni were stabbed on top of 0 . 35% Mueller Hinton Agar plates ( MHA Oxoid ) and incubated microaerobically at 42°C for 48 h . Autoagglutination ( AAG ) was performed as previously described [72] . Briefly the cells were harvested with 2 ml of phosphate-buffered saline ( PBS pH 7 . 2 ) and OD600 nm adjusted to 1 . 0 . The bacterial suspension was poured into sterile glass tubes ( 13 × 100 mm ) and incubated at 25°C , 37°C or 42°C for 24 h . 1 . 0 ml of the upper aqueous phase was carefully aspirated and the OD600 nm measured . Additionally viable bacteria were enumerated by plate counts . The lower 1 ml of solution containing the majority of the autoagglutinated cells were analysed by scanning electron microscopy ( SEM ) . Biofilm assays were performed according to Reeser et al [73] and Tram et al [74] . Assays were performed as previously described [20] with modifications . Approximately 100 µl of bacterial suspension containing 2 × 108–4 × 108 bacteria per ml was inoculated into 24-well plate containing a confluent monolayer of Caco-2 cells . A 5 min centrifugation step at 500 g was included to facilitate the movement of C . jejuni onto the surface of the cell monolayer . The cells were incubated for 45 min at 37°C in a 5% CO2 humidified atmosphere to allow passive adherence and internalization . For the adherence assay , the total number of bacteria associated with the cell layer was enumerated by viable count . For the invasion assay , infected cells were washed three times with PBS and 1 ml of MEM ( Dibco ) containing 400 µg/ml gentamicin was added to the cell monolayer for 3 h . The monolayer was washed three times with PBS and the cells lysed with 200 ml of 0 . 2% Triton X-100; intracellular bacterial counts were enumerated by viable count . Each assay was performed in triplicate . Cells were grown microaerobically for 24 h at 42°C , collected and washed three times with Brucella broth by centrifugation at 1000 g for 4 min . The final wash step was carried out using PBS and the OD600 nm adjusted to 0 . 025 . Bacteria were fixed on plastic cover slips with 2% glutaraldehyde and 5% formaldehyde solution for 10 min . Slides were washed three times with H2O and dehydrated in gradient steps of water/ethanol ( 15 , 30 , 50 , 75 , 90 and 100% ) . Last steps of dehydration were performed in 50% HMDS/ethanol solution ( Hexamethyldisilazane ) , followed by a final step of 100% HMDS . Slides were left to air dry and subsequently coated with gold ( 6 nm ) prior to analysis on Jeol 5000 Scanning Electron Microscope . Chicken colonisation analysis was performed as described previously [39] with an infective dose of 108 CFU ( BDD/02/11 ) . RNA was extracted using RNeasy kit according to the manufacturer's protocol ( Qiagen ) . cDNA synthesis and RT-qPCR was performed using 11168-O , Δtlp3 and Δtlp3c as previously described [33] . Nutrient depleted chemotaxis assay was performed as previously described [35] with the following modifications . One or two plugs ( 6 mm diameter ) equidistantly apart ( 60 mm ) were removed from Petri dishes containing 0 . 5% agar in H2O . Each well was filled with 0 . 5% agar containing 2 mM of selected amino acid . The plates were overlayed with 0 . 1% agar bacteriological ( Oxoid ) without nutritional supplements and left for 2 h to allow for diffusion of amino acids to create a chemical gradient . Cultures of C . jejuni were adjusted to OD600 nm 1 . 8 ( 109 cfu/ml ) . 100 µl drop of bacterial suspension was inoculated using a micropipette on top of the sloppy agar ( 0 . 1% agar bacteriological in H2O ) in the centre of the Petri dish with the plates incubated at 37°C for 4 h to allow chemotactic migration of the bacteria . To determine the number of viable bacteria associated with each amino acid plug , a 5 mm area around and including each plug was removed and placed into a microcentrifuge tube containing 900 µl of Brucella broth . These were incubated microaerobically for 1 h at 37°C to allow the bacteria to dissociate from the plug into the media . Viable counts were performed with enumeration by serial dilution followed by plate counts . C . jejuni 81116 flaA−/flaB− isogenic mutant was used as a non-motile , non-chemotactic control; additionally agar plugs containing no amino acid were used as a negative control . Yeast two-hybrid and three-hybrid analyses of protein interactions were performed as described previously [35] . | Bacterial chemotaxis is an important part in initiation of colonisation and subsequent pathogenicity . In this study , we report direct evidence supporting the involvement of C . jejuni transducer-like protein Cj1564 ( Tlp3 ) in the chemotaxis signalling pathway via small molecule arrays , surface plasmon and nuclear magnetic resonance ( SPR and NMR ) as well as chemotaxis assays of wild type and isogenic mutants . We further demonstrate its ability to interact with chemoattractants isoleucine , purine , malic acid and fumaric acid and chemorepellents lysine , glucosamine , succinic acid , arginine and thiamine . This is the first report identifying Cj1564 as a multi-ligand receptor for Campylobacter jejuni and its signal transduction initiation through the CheV and CheW proteins . Finally , our characterisation of C . jejuni Cj1564 provides additional basis for elucidating the roles of other group A chemoreceptors and their importance in the chemotaxis signalling pathway . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [] | 2014 | Characterisation of a Multi-ligand Binding Chemoreceptor CcmL (Tlp3) of Campylobacter jejuni |
Dipteran flies are amongst the smallest and most agile of flying animals . Their wings are driven indirectly by large power muscles , which cause cyclical deformations of the thorax that are amplified through the intricate wing hinge . Asymmetric flight manoeuvres are controlled by 13 pairs of steering muscles acting directly on the wing articulations . Collectively the steering muscles account for <3% of total flight muscle mass , raising the question of how they can modulate the vastly greater output of the power muscles during manoeuvres . Here we present the results of a synchrotron-based study performing micrometre-resolution , time-resolved microtomography on the 145 Hz wingbeat of blowflies . These data represent the first four-dimensional visualizations of an organism's internal movements on sub-millisecond and micrometre scales . This technique allows us to visualize and measure the three-dimensional movements of five of the largest steering muscles , and to place these in the context of the deforming thoracic mechanism that the muscles actuate . Our visualizations show that the steering muscles operate through a diverse range of nonlinear mechanisms , revealing several unexpected features that could not have been identified using any other technique . The tendons of some steering muscles buckle on every wingbeat to accommodate high amplitude movements of the wing hinge . Other steering muscles absorb kinetic energy from an oscillating control linkage , which rotates at low wingbeat amplitude but translates at high wingbeat amplitude . Kinetic energy is distributed differently in these two modes of oscillation , which may play a role in asymmetric power management during flight control . Structural flexibility is known to be important to the aerodynamic efficiency of insect wings , and to the function of their indirect power muscles . We show that it is integral also to the operation of the steering muscles , and so to the functional flexibility of the insect flight motor .
We undertook time-resolved microtomographic imaging of the thorax of tethered blowflies flying in the TOMCAT beamline of the Swiss Light Source [29] . We used single exposure phase retrieval to increase contrast by an order of magnitude over standard absorption-based imaging [30] . This was important to enable the high acquisition rates and short exposure times required to resolve the wingbeat cycle . The insects were tethered to a rotating stage that underwent four complete revolutions per recording , thereby allowing radiographs to be taken from multiple evenly spaced viewing angles whilst the insect was flying ( Figure 1 ) . We simultaneously captured the three-dimensional wingtip kinematics using stereo high-speed photogrammetry [31] and grouped the radiographs according to the wingtip position . Each group contained multiple radiographs corresponding to the same phase of the wingbeat , but taken from different viewing angles . This allowed us to reconstruct tomograms for each group separately , producing tomograms for ten evenly spaced phases of the wingbeat . Each tomogram pools radiographs from c . 600 wingbeats and therefore represents the average state of the flight motor at the corresponding phase of the wingbeat . The flies were rotated during radiographic acquisition ( 332° s−1 or 347° s−1 ) , producing a left-handed visual and inertial roll stimulus in the brightly lit lab environment ( Figures 1 , 2A , and 3 ) . The left wing had consistently higher stroke amplitude than the right wing ( 141±7° versus 100±9°; mean ± standard deviation ) , and a shallower stroke plane ( 47±4° versus 68±10° ) , typical of a stabilizing roll response [13] , [19] , [32] , [33] . The results of our experiments therefore allow us to compare the muscle strains and thoracic movements associated with simultaneous high versus low amplitude wingbeats in each individual . We analysed all three muscles inserting on the basalare sclerite ( b1 , b2 , b3 ) , and the two largest muscles ( I1 , III1 ) inserting on the first and third axillary sclerites ( Figures 4 and 5 ) . Together , these make up most of the mass of the steering muscles [6] , [7]; the other eight steering muscles are smaller and could not be distinguished reliably from the surrounding tissues . We first used our visualizations to describe the motions of the thoracic mechanisms that the steering muscles actuate ( Movie S1 , view here; Movie S2 , view here; Movie S3 , view here ) . The muscles that attach to the first axillary sclerite insert on its internal arm , which projects into the thorax and moves in opposition to the wing [7]; in contrast , the third axillary sclerite moves rather little relative to the base of the thorax ( Movie S2 , view here ) . The lever-like internal arm of the basalare sclerite oscillates back-and-forth ( Figure 6; Movie S2 , view here ) , while its external head articulates with a moving part of the thoracic wall called the pleural plate ( Figure 7; Movie S3 , view here ) . This hardened region of thoracic wall swings antero-ventrally on the downstroke , accommodated by the alternate opening and closing of two orthogonal clefts at its borders [34] . Rotation of the pleural plate was clearly responsible for driving oscillations of the basalare sclerite , which were of greater amplitude on the high-amplitude wing ( Movie S3 , view here ) . The wingbeat asymmetries that we measured were associated with bilateral asymmetries in steering muscle kinematics ( Figures 4–6; Movie S2 , view here ) , which we quantified by measuring strains directly from the tomograms ( Figures 5 , 6 , and 8 ) . We were unable to measure muscle resting length for the purposes of normalizing muscle strains because the flies were flying continuously and reacted to the roll stimulus throughout each recording . Instead , we referenced the strain of each muscle in a pair to the pooled mean length of both muscles , which allowed us to compare muscle strains within each pair and between flies . Mean muscle strain was bilaterally asymmetric within each muscle pair ( Figure 8C ) : higher on the high-amplitude wing for muscles I1 , III1 , and b3; but lower on the high-amplitude wing for muscles b1 and b2 ( Figure 6 ) . All of the muscles except III1 displayed detectable strain oscillations at wingbeat frequency , but we could only detect statistically significant bilateral amplitude asymmetries in muscles b1 and b3 ( Figure 8A ) . The amplitude of these strain oscillations was twice as high on the low-amplitude wing for b1 ( Figures 6C and 8A ) , and four times as high on the high-amplitude wing for b3 ( Figures 6A and 8A ) . The b1 strain oscillations also displayed a statistically significant phase asymmetry , with the oscillations on the low-amplitude wing delayed by a quarter of a wingbeat ( Figure 8B ) . Muscle strains need not always be caused by contraction of the muscle itself . For example , work-loop measurements have shown that b1 is specialized to do negative work ( i . e . , to absorb rather than impart kinetic energy ) , and is unable to cycle fast enough to drive oscillations at wingbeat frequency [20] . The measured b1 oscillations must therefore have been driven by oscillations of the basalare sclerite forced by movement of the wing and thorax ( Movie S1 , view here; Movie S2 , view here; Movie S3 , view here ) . We cannot say unequivocally why the b1 strain oscillations were bilaterally asymmetric , but in principle this must reflect either asymmetric loading or asymmetric stiffness . Electrophysiological studies have shown that b1 is activated earlier with increasing wingbeat amplitude , which increases both its stiffness and the amount of negative work done under a given strain [9] , [10] , [19] . It has therefore been hypothesised that this increased stiffness should cause the amplitude of the b1 muscle's oscillations to be lower when the wingbeat amplitude is higher . Our strain measurements support this hypothesis , but our visualizations show that the explanation is incomplete . This is because the lower amplitude oscillations of b1 on the high-amplitude wing are actually associated with larger oscillations of the basalare sclerite ( Figure 7; Movie S3 , view here ) . The picture is further complicated by the fact that b3 , which is expected to act antagonistically with b1 , also has higher amplitude oscillations on the high-amplitude wing ( Figure 6A ) . To resolve this puzzle , we examined the movements of the basalare sclerite in greater depth . Our visualizations show that movement of the basalare sclerite is dominated by rotation about its external head on the low-amplitude wing , but by dorso-ventral translation of the whole sclerite on the high-amplitude wing ( Movie S2 , view here; Movie S3 , view here ) . Consequently , the internal tip of the basalare sclerite traces an orbit that is aligned with b1 on the low-amplitude wing , but with b3 on the high-amplitude wing ( Figure 5C; Movie S2 , view here ) . These different modes of oscillation of the basalare sclerite explain why the strain amplitude is higher on the low-amplitude wing for b1 , but higher on the high-amplitude wing for b3 . We cannot determine how this is brought about , but one possibility is that the variable stiffness of the b1 muscle alters the impedance of the system anisotropically . Another possibility is that the orientation of the basalare sclerite is altered by the large b2 muscle [17] , [19] , which , like b1 , has a lower mean strain on the higher amplitude wing ( Figure 6D ) . Turning manoeuvres are associated with asymmetric aerodynamic power requirements , which cannot be met by varying the output of the power muscles asymmetrically [35] . We hypothesise that changing the mode of oscillation of the basalare sclerite serves to increase the amount of kinetic energy transferred to b1 on the low-amplitude wing , thereby absorbing excess muscle output . To test the plausibility of this hypothesis , we combined our measurements of b1 muscle strain with the results of a previous work-loop study [20] , to estimate the amount of negative work being done by b1 . Unlike the other steering muscles , b1 is typically active on both wings , although it is not necessarily activated on every wingbeat . We estimate that b1 would have done negative work at a rate of 0 . 04–0 . 06 mW on the high-amplitude wing ( 0 . 02 mW if inactive ) and 0 . 18–0 . 30 mW on the low-amplitude wing ( 0 . 06 mW if inactive ) . These intervals bracket the entire range of possible activation phase , and show that the b1 muscle could have been doing negative work at a rate up to 0 . 28 mW higher on the low-amplitude wing . This would be sufficient to manage anything up to a 24% asymmetry in the time-averaged aerodynamic power requirements of Calliphora , which have been estimated to be 1 . 58 mW per wing on the downstroke , and 0 . 81 mW per wing on the upstroke [36] . Our results therefore demonstrate that the b1 muscles could play a significant role in asymmetric power management , although it remains an open question whether the activation phase of b1 is controlled appropriately for this function . Our visualizations reveal a completely unexpected behaviour in another steering muscle , showing that the long tendon that connects the I1 muscle to the first axillary sclerite buckles when the wing is elevated above the wing hinge . This behaviour was observed on both wings in all four individuals , and was always greater on the high-amplitude wing ( Figure 9; Movie S2 , view here ) . Buckling only occurs under compressive loading , so it follows that both I1 muscles must be under compression in the upper part of the wingbeat . Consequently , I1 contraction cannot possibly increase stroke amplitude by exerting tensile stress on the first axillary sclerite at the top of the upstroke , contrary to what has been inferred previously from static anatomy [6] , [7] . Instead , I1 contraction must limit the movement of the wing at the bottom of the downstroke , thereby reducing stroke amplitude . Consistent with this interpretation , I1 muscle strain was always lower on the low-amplitude wing . This includes those points in the stroke cycle at which the tendon transitioned between its taut and buckled states . Since the I1 tendon must have been unloaded at these transition points , the fact that the muscle was shorter on the low-amplitude wing necessarily implies that I1 must have been contracted on the low-amplitude wing . This conclusion is consistent with the correlations observed in previous electrophysiological studies , which have found that I1 is only active at reduced stroke amplitude [8]–[10] . Buckling of the I1 tendon is important for two reasons . First , it accommodates higher amplitude movements of the first axillary sclerite than would otherwise be possible , because the effective strain measured along the straight line joining the origin of the tendon to the origin of I1 ( Figure 9C ) has four times the amplitude of the actual strain that the I1 muscle experiences on the high-amplitude wing ( Figure 6E ) . Second , it means that I1 contraction will always be intermittent in its effects within each stroke cycle , even if—like b1—the I1 muscle is unable to cycle at wingbeat frequency . Tendon buckling is not unique to I1 . Although we were unable to visualize the second muscle of the first axillary sclerite ( I2 ) fully , our visualizations show that the long tendon of this muscle also buckles on every wingbeat . Tendon buckling also occurs to a lesser extent in b3 ( Movie S2 , view here ) . This previously unknown phenomenon of tendon buckling may therefore be a rather general mechanism in the operation of the blowfly flight motor .
The fast , complex , three-dimensional movements of the insect flight motor are powered and controlled by several tens of linear actuators , each individually producing only a low-amplitude contractile strain . Here we have presented the first time-resolved visualisations of the workings of this extraordinary mechanism . Our results clearly show that the function of the steering muscles in controlling the wing kinematics can only be understood by placing them in the context of the deforming thoracic structures to which they attach . Deformations of the thoracic wall are not only responsible for transmitting forces from the power muscles to the wings , but are also important in accommodating qualitative changes in the modes of oscillation of the wing articulations . Likewise , deformations of the tendons connecting the steering muscles to the wing articulations are important in accommodating large excursions of the wing articulations , whilst permitting the steering muscles to curtail the wing's movement at certain stages of the stroke cycle . Structural flexibility is known to be important to the aerodynamic efficiency of insect wings [37] , and to the function of their indirect power muscles . We have now shown that it is integral also to the operation of the steering muscles , and so to the functional flexibility of the insect flight motor . We anticipate that the insights from this work will inspire the design of future micromechanical systems , and the technique that we have developed is of course applicable to other biological systems exhibiting periodic motion .
Blowflies ( C . vicina ) were collected from a permanent breeding colony at the Department of Bioengineering , Imperial College London and kept on a 24 h ( 12∶12 ) light-dark cycle . All individuals were used within two weeks of emergence at ambient lab temperature . Insects were cold-anesthetized at 4°C for 10 minutes and fixed dorsally by the scutum to a wooden tether , using a mixture of beeswax and colophonium . The scutum is a stiff , reinforced thoracic structure [7] , and is the standard mounting point for tethered flight preparations in flies . The wooden tethers were attached to a rotation stage using a custom-made holder to align the anteroposterior axis of the animals with the rotational axis of the end station ( Figure 1 ) . The insects ( n = 4 ) were placed in a 2 ms−1 airstream and left to settle into flight for >30 s before recording radiographs . The X-ray source was a superbending magnet located 25 m from the sample . Monochromatic and polychromatic beam configurations were available , and we ran experiments using both types of configuration for comparison . In the monochromatic configuration ( n = 2 ) , a double crystal multilayer monochromator was placed 7 m downstream of the source to extract monochromatic X-rays with a bandwidth of 2% at 18 keV photon energy ( wavelength = 0 . 7 Å ) and flux of 8×1011 ph/s−1 mm−2 at the sample site . The monochromator was removed in the polychromatic configuration ( n = 2 ) , which increased total photon flux by two orders of magnitude and increased the mean photon energy to 35 keV . However , the polychromatic beam was filtered to optimize the bandwidth and the peak wavelength value of the X-rays , which reduced the beam power to an estimated 2×1012 ph/s−1 mm−2 and mainly attenuated longer wavelengths . The beam was 10 mm wide and 4 . 1 mm high at the sample site under the monochromatic configuration , but was increased in height to 5 . 7 mm under the polychromatic configuration , which enabled visualization of the entire thorax ( Figure 2 ) . The polychromatic beam therefore offers the advantages of a higher flux and larger sampling volume compared to the monochromatic beam , but the algorithms used to reconstruct tomograms from the radiographs assume a specific beam energy , which is better defined for the monochromatic beam . In practice , we found no qualitative difference in the contrast or detail of the radiographs or tomograms between beam configurations , and conclude that both beam configurations allowed comparably good imaging . Results from both configurations are pooled in the analyses which follow . A 100 μm thick , Ce-doped LuAG scintillator was placed at a distance of 350 mm ( monochromatic configuration ) or 150 mm ( polychromatic configuration ) behind the sample to convert the transmitted X-rays into visible light . The scintillator distance was chosen to maximize the phase contrast of the radiographs and was dependent upon the mean photon energy ( 18 keV for the monochromatic beam and 35 keV for the polychromatic beam ) . The resulting edge-enhanced image was magnified using a custom-made , high numerical-aperture microscope ( Elya solutions , s . r . o ) offering continuously adjustable 2- to 4-fold magnification . Projection images were acquired with a pco . Dimax 12-bit CMOS detector system recording at 2 , 500 Hz for the monochromatic beam and 1 , 840 Hz for the polychromatic beam , while the insects were rotated at 347° s−1 or 332° s−1 , respectively . The laboratory environment provided a rich , high-contrast , visual scene , which would have stimulated the visual system of the insects strongly during rotation . The rotation rates of 347° s−1 and 332° s−1 were an order of magnitude higher than the lowest rates known to induce visually stimulated turning reactions in Diptera [9] , [14] . The angular velocity of the insect during rotation was three orders of magnitude lower than the mean wingtip velocity , so any bilateral asymmetries in the wing kinematics must have been due to changes in flight motor output in response to the roll stimulus , rather than passive aerodynamic effects due to rotation . Two synchronized Photron SA3 cameras ( Photron Ltd ) with 180 mm Sigma macro lenses were used to film the blowflies , recording at 4 , 000 Hz with a 33 . 3 μs exposure time and at 448×384 pixel image size ( Figure 1 ) . Illumination for the cameras was provided by a custom-built infrared LED light source directed onto white card below the insect . The cameras were calibrated using fully-automated calibration software running in Matlab ( The Mathworks Inc . ) [31] . We tracked the wingtips using background subtraction and manual thresholding to isolate the outlines of the wings in each camera view . The tip of each wing was determined as the point along the outline that was furthest from the wing hinge . The three-dimensional coordinates of the wingtip were then calculated using the camera calibration parameters . A data acquisition module ( National Instruments USB-6211 DAQ ) , sampling at 80 kHz , was used to record the exposure times of the Photron SA3 cameras and the pco . Dimax detector system for the purposes of grouping the radiographs . The flies had a mean wingbeat frequency of 145 Hz , so each 4 s recording consisted of approximately 600 wingbeats ( Figure 3 ) . We used the measured wingtip kinematics to group radiographs taken from different angles but at identical phases of the wingbeat . We identified the beginning and end of each wingbeat from the wingtip kinematics , and selected the radiographs closest in exposure time to ten evenly spaced phases of each wingbeat for analysis . This allowed us to combine data from all of the wingbeats measured for a given fly , despite the fact that their period was somewhat variable ( Figure 3 ) . Our tomographic reconstruction technique therefore produced one composite wingbeat for each individual , comprising ten time steps , where every time step pools radiographs from c . 600 wingbeats . The projections were despeckled to remove bright pixels caused by scattered X-rays hitting the detector , and were flat field corrected with the average flat-beam images ( i . e . , images taken with no sample ) and dark images ( i . e . , images taken with no beam ) acquired immediately after the scan . Phase retrieval was performed in a qualitative manner using the ANKAPhase implementation [38] single image phase retrieval algorithm under the assumption that the object consisted of a homogeneous soft tissue material [39] . We assumed that the steering muscles had a refractive index equal to that of water [40] . For the monochromatic beam , the real and imaginary parts of the deviation from one of the complex refractive index of the material were 7×10−7 and 5×10−10 , respectively . For the polychromatic beam , we assumed that the mean X-ray energy was 35 keV and used values of 2×10−7 and 10−10 for the real and imaginary parts , respectively , of the decrement from one of the index of refraction . Tomographic reconstruction was performed using a Fourier transform-based algorithm [41] . The resulting voxels had an isotropic spacing of 3 . 3 μm , with no discernible difference between tomograms collected using the monochromatic or polychromatic beam . The tomographic data were visualized and segmented using Amira ( VSG ) . We segmented the data using a manual threshold that separated the muscles and cuticle from the surrounding material ( Figure 5 ) . The manual threshold was chosen at a level approximately double that of the background noise ( Figure 10 ) . The end points of the muscles were manually tracked using natural features as markers to ensure that the same parts of the muscles were tracked from one frame to the next and between individuals ( Figure 5A ) . These end points were then used to calculate the lengths for each steering muscle ( Figure 5B ) . Both b3 and I1 exhibited tendon buckling during parts of the wingbeat . To take account of this , we used three-dimensional skeletonization [42] to find the line running through the centre of the tendon , which was then connected to the muscle ends to form a continuous line ( Figure 5B ) . A sinusoid of arbitrary mean , amplitude , and phase can be expressed as a linear combination of a sine function , a cosine function , and a constant . For each pair of steering muscles , we used a single linear model to regress the strains that we had measured for both wings on the sine and cosine of the wingbeat phase , comparing the fitted coefficients between wings . We did not control separately for fly identity , because the strain measurements had already been normalized by the mean value for each fly , such that the mean strain was the same for all flies ( i . e . , equal to zero ) . We used a Monte Carlo method to transform the 95% confidence intervals for the parameter estimates of the linear model into 95% confidence intervals for the mean , amplitude , and phase of the strain oscillations . This allowed us to test statistically for differences in the mean , amplitude , and phase of the strain oscillations between the high- and low-amplitude wings ( Figure 8 ) . Tu and Dickinson [20] measured the negative work done by the b1 muscle at different amplitudes of oscillatory strain , and with different phases of muscle activation , using the work loop technique . We interpolated their data to estimate the range of negative work that would be done by the muscle with the measured strain amplitudes of 2 . 3% and 5 . 5% . This allowed us to estimate that the net negative work done per wingbeat would have been in the range 0 . 25–0 . 41 μJ for the high-amplitude wing , and in the range 1 . 23–2 . 06 μJ for the low-amplitude wing , depending upon the unknown phase of muscle activation . The mean wingbeat frequency in our data was 145 Hz , so the b1 muscle would have been absorbing kinetic energy at a rate of 0 . 04–0 . 06 mW on the high-amplitude wing and 0 . 18–0 . 30 mW on the low-amplitude wing . If the muscle were inactive on either the high- or low-amplitude wing , kinetic energy would have been absorbed at a rate of 0 . 02 mW and 0 . 06 mW , respectively . Tethering is known to affect wing kinematics in other dipteran species [43] , but there is a paucity of free-flight data for Calliphora with which to compare our tethered wing kinematics , particularly during the roll manoeuvres that we have simulated . The mean wingbeat frequency ( 145±11 Hz ) and mean stroke plane angle on each wing ( 46 . 8°±4 . 1° low-amplitude wing , 68 . 0±9 . 6° high-amplitude wing ) , were within ranges observed in a free-flying Calliphora [44] , with similar wing length ( 9 . 2±0 . 5 mm free-flight data versus 8 . 7±0 . 4 mm in our data ) . Mean stroke amplitude on the high-amplitude wing ( 141°±7° ) was also within the range of free-flying Calliphora ( 123°–150° ) , but the mean stroke amplitude on the low-amplitude wing ( 100°±9° ) was slightly lower than previously recorded . However , free-flight kinematics have only been measured in symmetric flight conditions , and Calliphora typically reduce the stroke amplitude on the ipsilateral side during roll manoeuvres , rather than increasing it on the contralateral side , consistent with our measured kinematics [32] . Thus , we cannot discount an effect of tethering on our insects , but their wing kinematics appear to be broadly representative of those used during free-flight . A concern with using high-power X-rays to examine the biomechanics of the insect flight motor is that the radiation may affect the physiology of the insects during recording [45] . All four individuals continued flying after recording stopped , but although their measured wing kinematics fluctuated during recordings , there was no systematic change in the wing kinematics over the recording period ( Figure 3 ) . Stroke amplitude was bilaterally asymmetric throughout each recording , and was consistent with the asymmetry expected during a compensatory roll response , indicating that the flies were responsive throughout to the roll stimulus that we provided . Further evidence of the consistency of the flies' behaviour is provided by the quality of the tomograms themselves , because the tomographic reconstruction process will only be successful if the pose of the sample is consistent within each group of radiographs . Any significant variation in steering muscle kinematics between wingbeats would result in blurring of the reconstructed tomograms , which each represent the average state of the flight motor at a given phase of the wingbeat . The edge detail of the rigid scutum had similar edge sharpness to the steering muscles ( Figure 10 ) , which indicates that the steering muscle kinematics were consistent through each recording . Notwithstanding the consistency of their wing and muscle kinematics during the recordings , and the fact that the flies continued to fly immediately following exposure , all four individuals died a short while after . We therefore calculated the radiation dose received by the flies to assess the severity of exposure . Most of the X-rays produced by the beamline pass through the insects , but the amount will be dependent on both the individual ( due to variation in size and hydration ) and beam energy . We determined the proportion of X-rays absorbed by the insects by measuring the difference in image intensity between flat-beam images and radiographs where the insect was in the beam , using a region of interest containing the thorax , but not the mount . Using this method , we estimated that the mean absorption was 23% for the monochromatic beam and 13% for the polychromatic beam . The absorbed dose ( D ) was calculated as the absorbed power per unit mass:where a is the proportion of the beam absorbed by the insect , f is the beam flux , w is the width of the insect exposed to the beam ( estimated from the radiographs to be 3 . 3 mm ) , h is the height of the beam , m is the mass of the insect ( assumed to be 82 mg [20] ) , and t is the recording duration ( 4 s ) . The estimated total dose was 350 Gy for the monochromatic beam and 1 , 300 Gy for the polychromatic beam . These total doses are similar to or less than the doses that have been applied to other insects in previous work without any measurable long-term effect [45] . However , our dose rates ( 90 Gy s−1 and 325 Gy s−1 , for the monochromatic and polychromatic beam , respectively ) were at least an order of magnitude higher than those used in previous work [45] . We therefore attribute the adverse effects of radiation following exposure to the high rate at which the dose was supplied . | A blowfly's wingbeat is 50 times shorter than a blink of a human eye , and is controlled by numerous tiny steering muscles—some of which are as thin as a human hair . To visualize the movements of these muscles and the deformations of the surrounding exoskeleton , we developed a technique to allow us to look inside the insects during tethered flight . We used a particle accelerator to record high-speed X-ray images of the flying blowflies , which we used to reconstruct three-dimensional tomograms of their flight motor at ten different stages of the wingbeat . We measured the asymmetric movements of the steering muscles associated with turning flight , together with the accompanying movements of the wing hinge—arguably the most complex joint in nature . The steering muscles represent <3% of total flight muscle mass , so a key question has been how they can modulate the output of the much larger power muscles . We show that by shifting the flight motor between different modes of oscillation , the fly is able to divert mechanical energy into a steering muscle that is specialized to absorb mechanical energy . In general , we find that deformations of the muscles and thorax are key to understanding this remarkable mechanism . | [
"Abstract",
"Introduction",
"Conclusions",
"Materials",
"and",
"Methods"
] | [
"biotechnology",
"flight",
"mechanics",
"(biology)",
"x-radiation",
"engineering",
"and",
"technology",
"nervous",
"system",
"neuroscience",
"biological",
"locomotion",
"biomechanics",
"electromagnetic",
"radiation",
"materials",
"science",
"zoology",
"bioengineering",
"musculoskeletal",
"system",
"biomaterials",
"biophysics",
"physics",
"systems",
"biology",
"animal",
"physiology",
"bionics",
"anatomy",
"entomology",
"bone",
"and",
"joint",
"mechanics",
"biology",
"and",
"life",
"sciences",
"physical",
"sciences",
"motor",
"system"
] | 2014 | In Vivo Time-Resolved Microtomography Reveals the Mechanics of the Blowfly Flight Motor |
Human chromosome 14q32 . 2 harbors the germline-derived primary DLK1-MEG3 intergenic differentially methylated region ( IG-DMR ) and the postfertilization-derived secondary MEG3-DMR , together with multiple imprinted genes . Although previous studies in cases with microdeletions and epimutations affecting both DMRs and paternal/maternal uniparental disomy 14-like phenotypes argue for a critical regulatory function of the two DMRs for the 14q32 . 2 imprinted region , the precise role of the individual DMR remains to be clarified . We studied an infant with upd ( 14 ) pat body and placental phenotypes and a heterozygous microdeletion involving the IG-DMR alone ( patient 1 ) and a neonate with upd ( 14 ) pat body , but no placental phenotype and a heterozygous microdeletion involving the MEG3-DMR alone ( patient 2 ) . The results generated from the analysis of these two patients imply that the IG-DMR and the MEG3-DMR function as imprinting control centers in the placenta and the body , respectively , with a hierarchical interaction for the methylation pattern in the body governed by the IG-DMR . To our knowledge , this is the first study demonstrating an essential long-range imprinting regulatory function for the secondary DMR .
Human chromosome 14q32 . 2 carries a cluster of protein-coding paternally expressed genes ( PEGs ) such as DLK1 and RTL1 and non-coding maternally expressed genes ( MEGs ) such as MEG3 ( alias , GTL2 ) , RTL1as ( RTL1 antisense ) , MEG8 , snoRNAs , and microRNAs [1] , [2] . Consistent with this , paternal uniparental disomy 14 ( upd ( 14 ) pat ) results in a unique phenotype characterized by facial abnormality , small bell-shaped thorax , abdominal wall defects , placentomegaly , and polyhydramnios [2] , [3] , and maternal uniparental disomy 14 ( upd ( 14 ) mat ) leads to less-characteristic but clinically discernible features including growth failure [2] , [4] . The 14q32 . 2 imprinted region also harbors two differentially methylated regions ( DMRs ) , i . e . , the germline-derived primary DLK1-MEG3 intergenic DMR ( IG-DMR ) and the postfertilization-derived secondary MEG3-DMR [1] , [2] . Both DMRs are hypermethylated after paternal transmission and hypomethylated after maternal transmission in the body , whereas in the placenta the IG-DMR alone remains as a DMR and the MEG3-DMR is rather hypomethylated [1] , [2] . Furthermore , previous studies in cases with upd ( 14 ) pat/mat-like phenotypes have revealed that epimutations ( hypermethylation ) and microdeletions affecting both DMRs of maternal origin cause paternalization of the 14q32 . 2 imprinted region , and that epimutations ( hypomethylation ) affecting both DMRs of paternal origin cause maternalization of the 14q32 . 2 imprinted region , while microdeletions involving the DMRs of paternal origin have no effect on the imprinting status [2] , [5]–[8] . These findings , together with the notion that parent-of-origin specific expression patterns of imprinted genes are primarily dependent on the methylation status of the DMRs [9] , argue for a critical regulatory function of the two DMRs for the 14q32 . 2 imprinted region , with possible different effects between the body and the placenta . However , the precise role of individual DMR remains to be clarified . Here , we report that the IG-DMR and the MEG3-DMR show a hierarchical interaction for the methylation pattern in the body , and function as imprinting control centers in the placenta and the body , respectively . To our knowledge , this is the first study demonstrating not only different roles between the primary and secondary DMRs at a single imprinted region , but also an essential regulatory function for the secondary DMR .
We studied an infant with upd ( 14 ) pat body and placental phenotypes ( patient 1 ) and a neonate with upd ( 14 ) pat body , but no placental , phenotype ( patient 2 ) ( Figure 1 ) . Detailed clinical features of patients 1 and 2 are shown in Table 1 . In brief , patient 1 was delivered by a caesarean section at 33 weeks of gestation due to progressive polyhydramnios despite amnioreduction at 28 and 30 weeks of gestation , whereas patient 2 was born at 28 weeks of gestation by a vaginal delivery due to progressive labor without discernible polyhydramnios . Placentomegaly was observed in patient 1 but not in patient 2 . Patients 1 and 2 were found to have characteristic face , small bell-shaped thorax with coat hanger appearance of the ribs , and omphalocele . Patient 1 received surgical treatment for omphalocele immediately after birth and mechanical ventilation for several months . At present , she is 5 . 5 months of age , and still requires intensive care including oxygen administration and tube feeding . Patient 2 died at four days of age due to massive intracranial hemorrhage , while receiving intensive care including mechanical ventilation . The mother of patient 1 had several non-specific clinical features such as short stature and obesity . The father of patient 1 and the parents of patient 2 were clinically normal . We isolated genomic DNA ( gDNA ) and transcripts ( mRNAs , snoRNAs , and microRNAs ) from fresh leukocytes of patients 1 and the parents of patients 1 and 2 , from fresh skin fibroblasts of patient 2 , and from formalin-fixed and paraffin-embedded placental samples of patient 1 and similarly treated pituitary and adrenal samples of patient 2 ( although multiple body tissues were available in patient 2 , useful gDNA and transcript samples were not obtained from other tissues probably due to drastic post-mortem degradation ) . We also made metaphase spreads from leukocytes and skin fibroblasts . For comparison , we obtained control samples from fresh normal adult leukocytes , neonatal skin fibroblasts , and placenta at 38 weeks of gestation , and from fresh leukocytes of upd ( 14 ) pat/mat patients and formalin-fixed and paraffin-embedded placenta of a upd ( 14 ) pat patient [2] , [3] . We first examined the structure of the 14q32 . 2 imprinted region ( Figure 2 ) . Upd ( 14 ) was excluded in patients 1 and 2 as well as in the mother of patient 1 by microsatellite analysis ( Table S1 ) , and FISH analysis for the two DMRs identified a familial heterozygous deletion encompassing the IG-DMR alone in patient 1 and her mother and a de novo heterozygous deletion encompassing the MEG3-DMR alone in patient 2 ( Figure 2 ) . The microdeletions were further localized by SNP genotyping for 70 loci ( Table S1 ) and quantitative real-time PCR ( q-PCR ) analysis for four regions around the DMRs ( Figure S1A ) , and serial direct sequencing for the long PCR products harboring the deletion junctions successfully identified the fusion points of the microdeletions in patient 1 and her mother and in patient 2 ( Figure 2 ) . According to the NT_026437 sequence data at the NCBI Database ( Genome Build 36 . 3 ) ( http://preview . ncbi . nlm . nih . gov/guide/ ) , the deletion size was 8 , 558 bp ( 82 , 270 , 449–82 , 279 , 006 bp ) for the microdeletion in patient 1 and her mother , and 4 , 303 bp ( 82 , 290 , 978–82 , 295 , 280 bp ) for the microdeletion in patient 2 . The microdeletion in patient 2 also involved the 5′ part of MEG3 and five of the seven putative CTCF binding sites A–G [10] , and was accompanied by insertion of a 66 bp sequence duplicated from MEG3 intron 5 ( 82 , 299 , 727–82 , 299 , 792 bp on NT_026437 ) . Direct sequencing of the exonic or transcribed regions detected no mutation in DLK1 , MEG3 , and RTL1 , although several cDNA polymorphisms ( cSNPs ) were identified ( Table S1 ) . Oligoarray comparative genomic hybridization identified no other discernible structural abnormality ( Figure S1B ) . We next studied methylation patterns of the previously reported IG-DMR ( CG4 and CG6 ) and MEG3-DMR ( CG7 ) ( Figure 3A ) [2] , using bisulfite treated gDNA samples . Bisulfite sequencing and combined bisulfite restriction analysis using body samples revealed a hypermethylated IG-DMR and MEG3-DMR in patient 1 , a hypomethylated IG-DMR and differentially methylated MEG3-DMR in the mother of patient 1 , and a differentially methylated IG-DMR and hypermethylated MEG3-DMR in patient 2 , and bisulfite sequencing using placental samples showed a hypermethylated IG-DMR and rather hypomethylated MEG3-DMR in patient 1 ( Figure 3B ) . We also examined methylation patterns of the seven putative CTCF binding sites by bisulfite sequencing ( Figure 4A ) . The sites C and D alone exhibited DMRs in the body and were rather hypomethylated in the placenta ( Figure 4B ) , as observed in CG7 . Furthermore , to identify an informative SNP ( s ) pattern for allele-specific bisulfite sequencing , we examined a 349 bp region encompassing the site C and a 356 bp region encompassing the site D as well as a 300 bp region spanning the previously reported three SNPs near the site D , in 120 control subjects , the cases with upd ( 14 ) pat/mat , and patients 1 and 2 and their parents . Consequently , an informative polymorphism was identified for a novel G/A SNP near the site D in only a single control subject , and the parent-of-origin specific methylation pattern was confirmed ( Figure 4C ) . No informative SNP was found in the examined region around the site C , and no other informative SNP was identified in the two examined regions around the site D , with the previously known three SNPs being present in a homozygous condition in all the subjects analyzed . Finally , we performed expression analyses , using standard reverse transcriptase ( RT ) -PCR and/or q-PCR analysis for multiple imprinted genes in this region ( Figure 5A–5C ) . For leukocytes , weak expression was detected for MEG3 and SNORD114-29 in a control subject and the mother of patient 1 but not in patient 1 . For skin fibroblasts , although all MEGs but no PEGs were expressed in control subjects , neither MEGs nor PEGs were expressed in patient 2 . For placentas , although all imprinted genes were expressed in control subjects , PEGs only were expressed in patient 1 . For the pituitary and adrenal of patient 2 , DLK1 expression alone was identified . Expression pattern analyses using informative cSNPs revealed monoallelic MEG3 expression in the leukocytes of the mother of patient 1 ( Figure 5D ) , and biparental RTL1 expression in the placenta of patient 1 ( no informative cSNP was detected for DLK1 ) and biparental DLK1 expression in the pituitary and adrenal of patient 2 ( RTL1 was not expressed in the pituitary and adrenal ) ( Figure 5E ) , as well as maternal MEG3 expression in the control leukocytes and paternal RTL1 expression in the control placentas ( Figure S2 ) . Although we also attempted q-PCR analysis , precise assessment was impossible for MEG3 in the mother of patient 1 because of faint expression level in leukocytes and for RTL1 in patient 1 and DLK1 in patient 2 because of poor quality of mRNAs obtained from formalin-fixed and paraffin-embedded tissues .
The data of the present study are summarized in Figure 6 . Parental origin of the microdeletion positive chromosomes is based on the methylation patterns of the preserved DMRs in patients 1 and 2 and the mother of patient 1 as well as maternal transmission in patient 1 . Loss of the hypomethylated IG-DMR of maternal origin in patient 1 was associated with epimutation ( hypermethylation ) of the MEG3-DMR in the body and caused paternalization of the imprinted region and typical upd ( 14 ) pat body and placental phenotypes , whereas loss of the hypomethylated MEG3-DMR of maternal origin in patient 2 permitted normal methylation pattern of the IG-DMR in the body and resulted in maternal to paternal epigenotypic alteration and typical upd ( 14 ) pat body , but no placental , phenotype . In this regard , while a 66 bp segment was inserted in patient 2 , this segment contains no known regulatory sequence [11] or evolutionarily conserved element [12] ( also examined with a VISTA program , http://genome . lbl . gov/vista/index . shtml ) . Similarly , while no control samples were available for pituitary and adrenal , the previous study in human subjects has shown paternal DLK1 expression in adrenal as well as monoallelic DLK1 and MEG3 expressions in various tissues [11] . Furthermore , the present and the previous studies [2] indicate that this region is imprinted in the placenta as well as in the body . Thus , these results , in conjunction with the finding that the IG-DMR remains as a DMR and the MEG3-DMR exhibits a non-DMR in the placenta [2] , imply the following: ( 1 ) the IG-DMR functions hierarchically as an upstream regulator for the methylation pattern of the MEG3-DMR on the maternally inherited chromosome in the body , but not in the placenta; ( 2 ) the hypomethylated MEG3-DMR functions as an essential imprinting regulator for both PEGs and MEGs in the body; and ( 3 ) in the placenta , the hypomethylated IG-DMR directly controls the imprinting pattern of both PEGs and MEGs . These notions also explain the epigenotypic alteration in the previous cases with epimutations or microdeletions affecting both DMRs ( Figure S3 ) . It remains to be clarified how the IG-DMR and the MEG3-DMR interact hierarchically in the body . However , the present data , together with the previous findings in cases with epimutations [2] , [5]–[8] , imply that MEG3-DMR can remain hypomethylated only in the presence of a hypomethylated IG-DMR and is methylated when the IG-DMR is deleted or methylated irrespective of the parental origin . Furthermore , mouse studies have suggested that the methylation pattern of the postfertilization-derived Gtl2-DMR ( the mouse homolog for the MEG3-DMR ) is dependent on that of the germline-derive IG-DMR [13] . Thus , a preferential binding of some factor ( s ) to the unmethylated IG-DMR may cause a conformational alteration of the genomic structure , thereby protecting the methylation of the MEG3-DMR . It also remains to be elucidated how the IG-DMR and the MEG3-DMR regulate the expression of both PEGs and MEGs in the placenta and the body , respectively . For the MEG3-DMR , however , the CTCF binding sites C and D may play a pivotal role in the imprinting regulation . The methylation analysis indicates that the two sites reside within the MEG3-DMR , and it is known that the CTCF protein with versatile functions preferentially binds to unmethylated target sequences including the sites C and D [10] , [14]–[16] . In this regard , all the MEGs in this imprinted region can be transcribed together in the same orientation and show a strikingly similar tissue expressions pattern [1] , [12] , whereas PEGs are transcribed in different directions and are co-expressed with MEGs only in limited cell-types [1] , [17] . It is possible , therefore , that preferential CTCF binding to the grossly unmethylated sites C and D activates all the MEGs as a large transcription unit and represses all the PEGs perhaps by influencing chromatin structure and histone modification independently of the effects of expressed MEGs . In support of this , CTCF protein acts as a transcriptional activator for Gtl2 ( the mouse homolog for MEG3 ) in the mouse [18] . Such an imprinting control model has not been proposed previously . It is different from the CTCF protein-mediated insulator model indicated for the H19-DMR and from the non-coding RNA-mediated model implicated for several imprinted regions including the KvDMR1 [19] . However , the KvDMR1 harbors two putative CTCF binding sites that may mediate non-coding RNA independent imprinting regulation [20] , and the imprinting control center for Prader-Willi syndrome [21] also carries three CTCF binding sites ( examined with a Search for CTCF DNA Binding Sites program , http://www . essex . ac . uk/bs/molonc/spa . html ) . Thus , while each imprinted region would be regulated by a different mechanism , a CTCF protein may be involved in the imprinting control of multiple regions , in various manners . This imprinted region has also been studied in the mouse . Clinical and molecular findings in wildtype mice [1] , [22] , [23] , mice with PatDi ( 12 ) ( paternal disomy for chromosome 12 harboring this imprinted region ) [13] , [24] , [25] , and mice with targeted deletions for the IG-DMR ( ΔIG-DMR ) [22] , [26] and for the Gtl2-DMR ( ΔGtl2-DMR ) [27] are summarized in Table 2 . These data , together with human data , provide several informative findings . First , in both the human and the mouse , the IG-DMR is differentially methylated in both the body and the placenta , whereas the MEG3/Gtl2-DMR is differentially methylated in the body and exhibits non-DMR in the placenta . Second , the IG-DMR and the MEG3/Gtl2-DMR show a hierarchical interaction on the maternally derived chromosome in both the human and the mouse bodies . Indeed , the MEG3/Gtl2-DMR is epimutated in patient 1 and mice with maternally inherited ΔIG-DMR , and the IG-DMR is normally methylated in patient 2 and mice with maternally inherited ΔGtl2-DMR . Third , the function of the IG-DMR is comparable between human and mouse bodies and different between human and mouse placentas . Indeed , patient 1 has upd ( 14 ) pat body and placental phenotypes , whereas mice with the ΔIG-DMR of maternal origin have PatDi ( 12 ) -compatible body phenotype and apparently normal placental phenotype . It is likely that imprinting regulation in the mouse placenta is contributed by some mechanism ( s ) other than the methylation pattern of the IG-DMR , such as chromatin conformation [22] , [28] , [29] . Unfortunately , however , the data of ΔGtl2-DMR mice appears to be drastically complicated by the retained neomycin cassette in the upstream region of Gtl2 . Indeed , it has been shown that the insertion of a lacZ gene or a neomycin gene in the similar upstream region of Gtl2 causes severely dysregulated expression patterns and abnormal phenotypes after both paternal and maternal transmissions [30] , [31] , and that deletion of the inserted neomycin gene results in apparently normal expression patterns and phenotypes after both paternal and maternal transmissions [31] . ( In this regard , although a possible influence of the inserted 66 bp segment can not be excluded formally in patient 2 , phenotype and expression data in patient 2 are compatible with simple paternalization of the imprinted region . ) In addition , since the apparently normal phenotype in mice homozygous for ΔGtl2-DMR is reminiscent of that in sheep homozygous for the callipyge mutation [32] , a complicated mechanism ( s ) such as the polar overdominance may be operating in the ΔGtl2-DMR mice [33] . Thus , it remains to be clarified whether the MEG3/Gtl2-DMR has a similar or different function between the human and the mouse . Two points should be made in reference to the present study . First , the proposed functions of the two DMRs are based on the results of single patients . This must be kept in mind , because there might be a hidden patient-specific abnormality or event that might explain the results . For example , the abnormal placental phenotype in patient 1 might be caused by some co-incidental aberration , and the apparently normal placenta in patient 2 might be due to mosaicism with grossly preserved MEG3-DMR in the placenta and grossly deleted MEG3-DMR in the body . Second , the clinical features in the mother of patient 1 such as short stature and obesity are often observed in cases with upd ( 14 ) mat ( Table S2 ) . However , the clinical features are non-specific and appear to be irrelevant to the microdeletion involving the IG-DMR , because loss of the paternally derived IG-DMR does not affect the imprinted status [2] , [26] . Indeed , MEG3 in the mother of patient 1 showed normal monoallelic expression in the presence of the differentially methylated MEG3-DMR . Nevertheless , since the upd ( 14 ) mat phenotype is primarily ascribed to loss of functional DLK1 ( Figure S3B ) [2] , [34] , it might be possible that the microdeletion involving the IG-DMR has affected a cis-acting regulatory element for DLK1 expression ( for details , see Note in the legend for Table S2 ) . Further studies in cases with similar microdeletions will permit clarification of these two points . In summary , the results show a hierarchical interaction and distinct functional properties of the IG-DMR and the MEG3-DMR in imprinting control . Thus , this study provides significant advance in the clarification of mechanisms involved in the imprinting regulation at the 14q32 . 2 imprinted region and the development of upd ( 14 ) phenotype .
This study was approved by the Institutional Review Board Committees at National Center for Child health and Development , University College Dublin , and Dokkyo University School of Medicine , and performed after obtaining written informed consent . All the primers utilized in this study are summarized in Table S3 . For leukocytes and skin fibroblasts , genomic DNA ( gDNA ) samples were extracted with FlexiGene DNA Kit ( Qiagen ) , and RNA samples were prepared with RNeasy Plus Mini ( Qiagen ) for DLK1 , MEG3 , RTL1 , MEG8 and snoRNAs , and with mirVana miRNA Isolation Kit ( Ambion ) for microRNAs . For paraffin-embedded tissues including the placenta , brain , lung , heart , liver , spleen , kidney , bladder , and small intestine , gDNA and RNA samples were extracted with RecoverAll Total Nucleic Acids Isolation Kit ( Ambion ) using slices of 40 µm thick . For fresh control placental samples , gDNA and RNA were extracted using ISOGEN ( Nippon Gene ) . After treating total RNA samples with DNase , cDNA samples for DLK1 , MEG3 , MEG8 , and snoRNAs were prepared with oligo ( dT ) primers from 1 µg of RNA using Superscript III Reverse Transcriptase ( Invitrogen ) , and those for microRNAs were synthesized from 300 ng of RNA using TaqMan MicroRNA Reverse Transcription Kit ( Applied Biosystems ) . For RTL1 , cDNA samples were synthesized with RTL1-specific primers that do not amplify RTL1as . Control gDNA and cDNA samples were extracted from adult leukocytes and neonatal skin fibroblasts purchased from Takara Bio Inc . Japan , and from a fresh placenta of 38 weeks of gestation . Metaphase spreads were prepared from leukocytes and skin fibroblasts using colcemide ( Invitrogen ) . Microsatellite analysis and SNP genotyping were performed as described previously [2] . For FISH analysis , metaphase spreads were hybridized with a 5 , 104 bp FISH-1 probe and a 5 , 182 bp FISH-2 probe produced by long PCR , together with an RP11-566I2 probe for 14q12 used as an internal control [2] . The FISH-1 and FISH-2 probes were labeled with digoxigenin and detected by rhodamine anti-digoxigenin , and the RP11-566I2 probe was labeled with biotin and detected by avidin conjugated to fluorescein isothiocyanate . For quantitative real-time PCR analysis , the relative copy number to RNaseP ( catalog No: 4316831 , Applied Biosystems ) was determined by the Taqman real-time PCR method using the probe-primer mix on an ABI PRISM 7000 ( Applied Biosystems ) . To determine the breakpoints of microdeletions , sequence analysis was performed for long PCR products harboring the fusion points , using serial forward primers on the CEQ 8000 autosequencer ( Beckman Coulter ) . Direct sequencing was also performed on the CEQ 8000 autosequencer . Oligoarray comparative genomic hybridization was performed with 1×244K Human Genome Array ( catalog No: G4411B ) ( Agilent Technologies ) , according to the manufacturer's protocol . Methylation analysis was performed for gDNA treated with bisulfite using the EZ DNA Methylation Kit ( Zymo Research ) . After PCR amplification using primer sets that hybridize both methylated and unmethylated clones because of lack of CpG dinucleotides within the primer sequences , the PCR products were digested with appropriate restriction enzymes for combined bisulfite restriction analysis . For bisulfite sequencing , the PCR products were subcloned with TOPO TA Cloning Kit ( Invitrogen ) and subjected to direct sequencing on the CEQ 8000 autosequencer . Standard RT-PCR was performed for DLK1 , RTL1 , MEG3 , MEG8 , and snoRNAs using primers hybridizing to exonic or transcribed sequences , and one µl of PCR reaction solutions was loaded onto Gel-Dye Mix ( Agilent ) . Taqman real-time PCR was carried out using the probe-primer mixtures ( assay No: Hs00292028 for MEG3 and Hs00419701 for MEG8; assay ID: 001028 for miR433 , 000452 for miR127 , 000568 for miR379 , and 000477 for miR154 ) on the ABI PRISM 7000 . Data were normalized against GAPDH ( catalog No: 4326317E ) for MEG3 and MEG8 and against RNU48 ( assay ID: 0010006 ) for the remaining miRs . The expression studies were performed three times for each sample . To examine the imprinting status of MEG3 in the leukocytes of the mother of patient 1 , direct sequence data for informative cSNPs were compared between gDNA and cDNA . To analyze the imprinting status of RTL1 in the placental sample of patient 1 and that of DLK1 in the pituitary and adrenal samples of patient 2 , RT-PCR products containing exonic cSNPs informative for the parental origin were subcloned with TOPO TA Cloning Kit , and multiple clones were subjected to direct sequencing on the CEQ 8000 autosequencer . Furthermore , MEG3 expression pattern was examined using leukocyte gDNA and cDNA samples from multiple normal subjects and leukocyte gDNA samples from their mothers , and RTL1 expression pattern was analyzed using gDNA and cDNA samples from multiple fresh normal placentas and leukocyte gDNA from the mothers . | Genomic imprinting is a process causing genes to be expressed in a parent-of-origin specific manner—some imprinted genes are expressed from maternally inherited chromosomes and others from paternally inherited chromosomes . Imprinted genes are often located in clusters regulated by regions that are differentially methylated according to their parental origin . The human chromosome 14q32 . 2 imprinted region harbors the germline-derived primary DLK1-MEG3 intergenic differentially methylated region ( IG-DMR ) and the postfertilization-derived secondary MEG3-DMR , together with multiple imprinted genes . Perturbed dosage of these imprinted genes , for example in patients with paternal and maternal uniparental disomy 14 , causes distinct phenotypes . Here , through analysis of patients with microdeletions recapitulating some or all of the uniparental disomy 14 phenotypes , we show that the IG-DMR acts as an upstream regulator for the methylation pattern of the MEG3-DMR in the body but not in the placenta . Importantly , in the body , the MEG3-DMR functions as an imprinting control center . To our knowledge , this is the first study demonstrating an essential function for the secondary DMR in the regulation of multiple imprinted genes . Thus , the results provide a significant advance in the clarification of underlying epigenetic features that can act to regulate imprinting . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"genetics",
"and",
"genomics/epigenetics",
"genetics",
"and",
"genomics/genetics",
"of",
"disease",
"genetics",
"and",
"genomics/medical",
"genetics"
] | 2010 | The IG-DMR and the MEG3-DMR at Human Chromosome 14q32.2: Hierarchical Interaction and Distinct Functional Properties as Imprinting Control Centers |
Host genetic variation modifying HIV-1 acquisition risk can inform development of HIV-1 prevention strategies . However , associations between rare or intermediate-frequency variants and HIV-1 acquisition are not well studied . We tested for the association between variation in genic regions and extreme HIV-1 acquisition phenotypes in 100 sub-Saharan Africans with whole genome sequencing data . Missense variants in immunoglobulin-like regions of CD101 and , among women , one missense/5’ UTR variant in UBE2V1 , were associated with increased HIV-1 acquisition risk ( p = 1 . 9x10-4 and p = 3 . 7x10-3 , respectively , for replication ) . Both of these genes are known to impact host inflammatory pathways . Effect sizes increased with exposure to HIV-1 after adjusting for the independent effect of increasing exposure on acquisition risk . Trial registration: ClinicalTrials . gov NCT00194519; NCT00557245
The discovery of the protective deletion variant , CCR5-delta32 , in the chemokine receptor 5 gene , encoding a HIV-1 co-receptor [1–3] , generated great enthusiasm to search for additional host genetic variants and pathways associated with HIV-1 acquisition as a means of identifying targets for new HIV-1 prevention and treatment strategies . This enthusiasm was further enhanced by documentation of HIV-1 exposed seronegative ( HESN ) individuals who had very high exposure and lacked CCR5-delta32 , [4–7] suggesting existence of additional genetic factors that alter the risk of sexually transmitted HIV-1 infection ( Online Mendelian Inheritance in Man [OMIM] phenotype #609423 ) [8] . At least one in vitro experiment supports the hypothesis of a strong genetic component to infection risk , reporting 50% heritability in cellular susceptibility to HIV-1 infection [9] . Nevertheless , genome-wide association studies ( GWAS ) to date searching for such genetic risk factors for HIV-1 infection risk have met with limited success [10–17] . Most GWAS have had moderate ( ~80% ) average power to detect common variants , very low power to detect variants with minor allele frequency ( MAF ) = 5% ( approximately 1% power for an OR = 2 ) and even less power to detect associated rare variants ( RVs ) ( MAF≤1% ) . Additionally , susceptibility to HIV-1 can only be assessed among individuals who are exposed to the virus . While some assessment of HIV-1 exposure was used for most HIV-1 GWAS [10–17] , exposure measurement error and/or exposure misclassification , including that related to lack of information about the HIV-1 infected partners’ plasma HIV-1 RNA level ( s ) , can result in lower statistical power than anticipated . Furthermore , out of necessity , some studies have been forced to attempt replication across different ancestral/racial groups [10] . However , risk variants might differ between such groups , thereby lessening the power for replication . Hence , major gaps in HIV-1 genetic association studies still exist , and focus on power to detect rare associated variants employing high accuracy in HIV-1 exposure measurements is warranted . The contribution of rare variants to risk of HIV-1 acquisition is of particular interest because effect size is generally inversely correlated with MAF when an association does exist [18] and large effects can be expected to translate to strong interventional impact ( e . g . , HMG-CoA reductase inhibitors and familial hypercholesterolemia [19] ) . With these issues in mind , we undertook an association study of HIV-1 acquisition using whole genome sequencing ( WGS ) of extreme phenotypes sampled from two large clinical trials and one observational study of African HIV-1 serodiscordant couples ( stable heterosexual couples with one partner HIV-1-infected and the other partner HIV-1-seronegative at enrollment ) ( n = 8 , 593 couples ) . These studies ( S1 Table ) had similar clinical follow-up , including quarterly risk assessments , PCR-verification of HIV-1 infection , reports of protected and unprotected sexual activity from both partners , measurement of the infected partner’s plasma HIV-1 RNA level and molecular confirmation of transmission linkage through viral sequencing [20–22] .
The 100 Discovery stage genomes were sequenced by Complete Genomics Inc . ( CGI ) with high quality results ( S2 Table ) . The RVT1 test ( “rare variant test 1” ) [24] , a statistical test designed specifically for rare variant association studies , was used to test the difference between extremes in functional variant burden ( defined as the total number of minor alleles ( aggregated variant scores ) by gene comparing cases and controls; see Materials and methods ) for each of 18 , 354 genic regions , including 284 , 632 functional variants in the tests . The regions with the two lowest RVT1 p-values ( S2A Fig ) had an estimated 83% probability that at least one of these was a true positive , based on a False Discovery Rate ( FDR ) analysis [25] . QQ-plots of the RVT1 results showed good adherence to expected behavior and no evidence of confounding by major ancestry nor by spatially-isolated pockets of ancestry [26] ( S2B and S2C Fig ) . These two regions are transcribed regions of CD101 ( NCBI Gene ID: 9398 ) and UBE2V1 ( NCBI Gene ID: 7335 ) . For both of these genes , individuals with greater numbers of polymorphic functional sites had increased risk of HIV-1 acquisition: CD101 odds ratio ( OR ) = 2 . 7 ( per functional site with at least one minor allele ) , 95% CI = [1 . 6–4 . 8] , p = 3 . 6x10-5 , and UBE2V1 OR = 3 . 7 , 95% CI = [1 . 8–7 . 5] , p = 4 . 7x10-5 ( Table 2 ) . Within these two genes in the Discovery stage , eight variants were novel at the time of identification , and all eight validated by Sanger sequencing ( S3 Table ) . Based on the FDR results showing >80% chance of a true positive , the CD101 and UBE2V1 regions were moved forward to the Replication stage for testing in a longitudinal analysis . Test regions from the Discovery stage almost certainly include variants significantly associated , as well as , variants not significantly associated ( i . e . , noise ) with outcome . The benefits , and perhaps even necessity , of using biological knowledge to increase signal-to-noise ratio for rare variant replication is well recognized [27 , 28] . To increase the signal-to-noise ratio and increase power , we prioritized and grouped variants ( see Materials and methods ) in CD101 and UBE2V1 for Replication stage testing . Fourteen functional variants in CD101 were designated as “primary replication variants” ( PRVs ) based on a direction of effect consistent with that for the Discovery result ( S4 Table ) and their degree of significance in by-variant tests ( S3A Fig ) . These variants were subdivided into four sub-groups for replication testing ( see Materials and methods ) : ( 1 ) five missense variants in regions encoding extracellular CD101 immunoglobulin-like ( Ig-like ) protein domains [29] , ( 2 ) five missense variants in the CD101 cytoplasmic domain , ( 3 ) two 3’-UTR variants and ( 4 ) two splice site variants ( Fig 1A , S3A Fig , S4 Table ) . Creating four separate a priori replication test groups increases the multiple-testing penalty , but is expected to further increase the signal-to-noise ratio within some of these variant test groups . Similarly , 11 of 15 predicted functional variants in UBE2V1 were designated as PRVs , and these were divided into two groups: ( 1 ) six 5’-UTR and ( 2 ) five 3’-UTR variants ( Fig 1B , S3B Fig , S5 Table ) . One of the 5’-UTR variants , rs6095771 , is identified as a missense variant in the canonical transcript , and as a 5’-UTR for other transcripts . It was grouped with the 5’-UTR variants due to low predicted power to replicate a single variant with MAF = 0 . 02 . Three of these variants were novel and another six are not found among Kenyans in the 1kG database [30] ( S5 Table ) . Furthermore , two of the functional UBE2V1 Discovery variants that are indexed in the ESP [31] or ExAC [32] databases are extremely rare and found in ESP only among individuals of African descent ( rs187204768 , MAF = 4/4046 and rs6095771 , MAF = 20/4400 ) ( S5 Table ) . These results indicate that the UBE2V1 association could not be detected at the variant level using GWA-type methods and can only be detected in a population with substantial recent African ancestry , such as the present study . Across these three HIV-1 serodiscordant couples cohorts , the pool of participants who were available for the Replication stage had varying amounts of reported sexual exposure . Because inclusion of individuals with no or little HIV-1 exposure could reduce power , we identified and excluded these individuals using a Protected-sex Index ( PI ) . PI is defined as the proportion of study visits for which only abstinence or 100% condom use was reported; we considered this our measure of baseline behavior tendency . PI predicted overall HIV-1 seroconversion rates among the three HIV-1 cohorts ( Fig 2 ) , indicating reasonably low measurement error and a high signal-to-noise ratio for PI . Simulation studies predicted that power would be maximized when only individuals with PI ≤0 . 6 were included in the Replication analysis , despite a larger sample size when individuals with lower exposure are included ( see Materials and methods ) . Accordingly , the Replication cohort was restricted to the 261 individuals HIV-1 uninfected at baseline , who each had PI ≤ 0 . 6 and were not selected for the Discovery stage analysis . Genotyping of variants in CD101 and UBE2V1 for the 261 Replication stage individuals was completed using molecular inversion probe sequencing ( MIPs ) [33] technology ( see Materials and methods ) . Given the cost efficiency of using MIPs , an auxiliary sample of 968 participants selected by more common methods rather than sexual exposure also was sequenced and used for verification of simulated power studies ( see Materials and methods ) for a total of 1229 individuals with MIPs data for CD101 and UBE2V1 ( S6 Table ) . Neither CD101 splice variant from the Discovery stage was found in the Replication sample ( S7A Table ) , reducing the number of CD101 primary test groups from four to three . In total , 83 CD101 SNVs were detected among the 1229 individuals ( S8 Table ) . For UBE2V1 , three of five 3’-UTR PRVs were found in the Replication sample but only one of the six 5’-UTR PRVs was present ( S7B & S9 Tables ) . This is consistent both with the rarity of these UBE2V1 variants and enrichment for these variants in the Discovery sample . To test for further evidence of a true positive association between HIV-1 acquisition risk in women and Ig-like CD101 variants or UBE2V1 rs6095771 , we augmented the Replication data with variant data from the auxiliary sample and tested for an interaction between the reported PI score and each PRV score to ascertain whether a dose-response relationship was present . “Dose” here is HIV-1 virus exposure quantified in terms of frequency of unprotected sex with a partner with the average plasma HIV-1 RNA in the Replication sample partners . For UBE2V1 rs6095771 , this increased the number of women with 5’-UTR PRVs in the analysis from 4 to 28 ( S7B Table ) . We found strong dose-response relationships for both genes , indicating that the association between variant scores and HIV-1 risk is positively related to the frequency of unprotected sex: p = 5 . 0x10-8 for the CD101 Ig-like dose-response and p = 8 . 2x10-6 for the UBE2V1 rs6095771 dose-response , with exposure assessed through the PI score and modeled on a log-scale ( Fig 4; S7 Fig ) . Because the model is adjusted for PI , these significant associations with increasing dose are in addition to the effect of decreasing PI . Overall model significance levels including the variant and dose-response variables were marked at p = 5 . 1x10-12 and p = 1 . 7x10-12 , respectively , for CD101 and UBE2V1 in explaining variation in HIV-1 seroconversion risk ( Tables 4 and 5 ) . The interaction models also provide estimated HRs for the risk groups under the assumptions of no protection and the average frequency of heterosexual intercourse among those in the Replication Stage ( CD101 HR = 5 . 4 , p = 1 . 4x10-3 , 95% CI = [1 . 9 , 15 . 2]; and UBE2V1 HR = 16 . 2 , p = 2 . 8x10-4 , 95% CI = [3 . 6 , 72 . 6] ) for the Ig-like PRV score and UBE2V1 rs6095771 , respectively . Again , these HRs are adjusted for PI , which means that the increase in effect size is above that explained by the PI variable . When we included all 1229 samples genotyped after the Discovery stage ( Replication sample and auxiliary sample ) in a model that does not account for sexual exposure/PI we found highly attenuated HRs ( Tables 4 and 5 ) and non-significant p-values . This phenomenon occurs because the estimated HRs are averages over a group that contains individuals who have no risk due to no sexual exposure , “diluting” the effect we detected in those with higher exposure . To illustrate these effects as a function of the exposure to HIV-1 viral quantity , HRs were estimated with step-wise decreases in sample size , excluding exposed individuals at each step ( Fig 4 , S6 & S7 Figs ) . Levels of cytokines in blood are a useful measure for immunologic function and may indicate presence of a generalized host pro-inflammatory state [36] . We evaluated whether the three CD101 Ig-like PRVs with FDR < 0 . 05 in the replication stage were also associated with altered plasma cytokine levels compared to individuals without any of these variants ( see Materials and methods ) . Among the 163 individuals in this subset for whom CD101 genotypes were determined and for whom measurements of 25 plasma cytokines were available [37] , carriers of these CD101 risk alleles had significantly lower serum levels of IL1R1 ( right-shifted distribution ) compared to those without any CD101 risk alleles ( OR = 0 . 19 for achieving the 75th percentile IL1R1 value , 95% CI = [0 . 07 , 0 . 54] , p = 1 . 7x10-3; adjusted p = 0 . 04 ) ( S8 and S9 Figs ) . There was also a tendency toward lower levels of sCD40L ( p = 0 . 0049; S10 Table ) . Frequencies of the UBE2V1 5’-UTR PRVs were too low in the individuals with cytokine measurements to allow for assessment of any association with cytokine levels . None of these variants were associated with HIV-1 plasma RNA set point among HIV-1 seroconverters ( S11 Table ) , indicating that the associations we have found are unlikely to act through altered viral replication with these gene products functioning as intracellular viral restriction factors [38] .
Our findings demonstrate that aggregates of host genetic variants , including variants with MAF<10% , can have a strong and replicated association with HIV-1 acquisition risk . Our replication of CD101 Ig-like variants showed an HR for HIV-1 infection of 4 . 3 ( 95% CI = [2 . 1–8 . 9] , p = 6 . 4x10-5 ) ; and replication of association of UBE2V1 rs6095771 with HIV-1 acquisition risk showed an HR of 6 . 4 in women ( 95% CI = [2 . 1 , 19 . 1] , p = 9 . 5x10-4 ) . Two CD101 missense/regulatory variants reached individual statistical significance after adjustment for multiple testing , four had individual FDRs < 0 . 05 , and a strong dose-response relationship with the frequency of unprotected sex collectively strengthened evidence that association with HIV-1 acquisition is real . While many of these variants were individually rare or infrequent , nearly 18% of Kenyans evaluated in the 1K Genomes Project [30] had one or more of the three most common CD101 Ig-like HIV-1 risk variants ( rs17235773 , rs3754112 , rs12093834 ) . Hence , these variants or others in CD101 and UBE2V1 could have a substantial impact on the population-based HIV-1 infection risk . CD101 and UBE2V1 are both biologically plausible candidates for influencing HIV-1 acquisition . CD101 is expressed on CD4+ and CD8+ T-cells , dendritic cells and monocytes , [39] and appears to alter CD4+/CD25+/FOXP3+ T regulatory cell ( Treg ) function based on both a murine graft-versus-host disease model [40] , and through IL10 secretion from human dendritic cells [41] . Monoclonal antibody ligation of CD101 reduces T cell proliferation through a Ca2+ and tyrosine kinase-dependent pathway possibly by preventing translocation of nuclear factor of activated T cells ( NFAT ) and IL-2 production [42] . Recent experiments using a murine model of chronic colitis demonstrate that adoptive transfer of CD101-/- Tregs is associated with Th17 cell proliferation and more severe colitis [43] . CD101 expression is also strongly associated with the immune suppression function of Tregs in humans . [44] Reduced expression of CD101 on mucosal CD8+ T cells has been associated with increased tissue inflammation in studies of human intestinal [45] , and pulmonary mucosa . [46] Given that local inflammation and CD4+ [47] and CD8+ [48 , 49] T cell immune activation have been associated with increased HIV-1 acquisition risk , and reduced immune activation [50 , 51] or immune quiescence [52 , 53] have been associated with natural resistance to HIV-1 in HESN , our results support the idea that CD101 gene variants may modify HIV-1 heterosexual acquisition risk through altered levels of genital mucosal inflammation . The finding that CD101 risk variants are associated with lower plasma IL1R1 levels , but not with HIV-1 RNA set point , suggests that these variants have systemic immunological effects in the seronegative partner while not directly acting on HIV-1 replication . Recent studies in mouse models indicate that IL1 may inhibit Treg and enhance Th17 differentiation [54] . However , it is unclear at this point how variants in CD101 , specifically those identified in CD101 Ig-like domains , might modify either IL1 or IL1R1 levels . The IL1R1 rs2234650 genotype has been reported to be associated with HIV-1 acquisition in infants of HIV-1 infected mothers with modification of risk by IL1 gene family haplotypes [55] . In addition to CD101 being associated with increased Treg function , IL1R1 expression has been associated with increased Treg and anti-inflammatory IL10 secretion [44] . Combining these prior data with our findings , we hypothesize that CD101 Ig-like variants reduce Treg function with associated reductions in IL1R1 levels and an enhanced pro-inflammatory environment , which leads to increased risk of HIV-1 acquisition . While efforts to test this hypothesis and to develop a more detailed understanding of CD101 function are underway , our results suggest that targeting CD101 activity could be a novel approach to host-directed , HIV-1 prevention . UBE2V1 associates with TRIM5-α , a host restriction factor involved with HIV-1 capsid uncoating [56]; however , rare and uncommon UBE2V1 variants have not previously been studied in association with HIV-1 acquisition risk . Previous GWAS without rare variant burden/aggregation tests cannot adequately assess the UBE2V1 association observed here because of the rarity of the variants found to be associated with HIV-1 acquisition risk in this study . UBE2V1 forms an ubiquitin-conjugating complex generating unattached polyubiquitin chains that may stimulate NF-κB activation and consequent pro-inflammatory cytokine production [56 , 57] . This complements reports of reduced systemic immune activation associated with resistance to HIV-1 acquisition in Kenyan sex workers [50 , 53 , 58] . Of note , UBE2V1 is one of only 23 genes differentially down-regulated by HIV-1 trans-activator of transcription ( TAT ) , an HIV-1 protein that is required for efficient replication of the HIV-1 virus and potential escape from the host immune system [59] . Distinguishing between a group carrying variants protecting against HIV-1 acquisition versus a group carrying risk-increasing variants requires that both groups are exposed to HIV-1 , i . e . to assess susceptibility to HIV-1 in any given individual , that individual must be exposed to HIV-1 . This translates directly to a mathematical proof that statistical power to detect a significant association with HIV-1 acquisition risk is increased by selectively identifying individuals with sustained high levels of HIV-1 exposure . Indeed , the success of this study was dependent on our ability to quantify HIV-1 exposure with relatively high accuracy to identify exposed individuals . Given that the overall per-contact probability of heterosexual HIV-1 transmission is intrinsically low ( estimated at ~1/1000 for vaginal intercourse among the study population ) [60] , our findings suggest that inclusion of individuals with lesser HIV-1 exposure may explain , in part , why it has been difficult to identify and/or validate genetic risk factors for HIV-1 acquisition . The use of the extreme phenotypes design is another strength of this study . The discovery of CCR5-delta32 was based on the observation that extreme resistance phenotypes existed [61 , 62] leading to case-control studies to identify the CCR5-delta32 variant among candidate genes [1–3] , including extreme hemophiliac controls [1] and the others employing just two to four extreme resistant controls [2 , 3] . Identifying individuals with phenotypes that represent the extremes of risk of HIV-1 acquisition has been challenging because few studies are able to assess both the behavioral and biologic dimensions ( e . g . frequency of unprotected sex , and plasma HIV-1 RNA level in the HIV-1 transmitting partner ) that contribute to exposure . Our use of data from both partners added accuracy to the phenotypes we used in this study—e . g . , plasma HIV-1 RNA level contributed by the HIV-1 infected partner , reported frequency of unprotected sex from both partners , and epidemiologic data from the HIV-1 uninfected partner ( e . g . , male circumcision status ) [23] . Further improvements in accurate exposure quantification could increase power even more and perhaps very extreme phenotypes can be identified that provide superb power with small samples but lead to generalizable treatments . Recent examples that underscore the value of the extreme phenotype analysis approach include discovery using a sample size of a dozen individuals of a human antibody to a malaria protein that prevents death and provides a promising new malaria vaccine target [63] , and discovery of the protective effect of PKC9 loss of function variants against cardiovascular disease in a small group of extreme individuals [64] , leading to the development of PKC9 inhibitors for lowering LDL cholesterol . Our analysis identified variants in CD101 and UBE2V1 associated with increased risk of HIV-1 acquisition , but we did not identify variants associated with reduced risk of HIV-1 acquisition . Statistical power is lower to identify protective variants than it is to identify risk variants ( of the same magnitude but inverse ) when the outcome has low incidence . Discovery of protective variants for HIV-1 infection in a population with average sexual contact and some use of protection against transmission would require extended observation time to detect differences in the survivor rates ( “rates of non-seroconversion , ” ) while differences in rates of seroconversion of the same magnitude can be detected statistically within a shorter time period . A limitation of these results is that confounding cannot be ruled out with certainty as the source of the associations . This is true for any observational study , though observational studies remain a fundamental part in building a step-wise scientific case for causal association for many exposures that cannot be tested experimentally in humans ( including exposure to genetic variants , the epidemiological exposure being tested here ) , with smoking as a cause of lung cancer being a prime example . [65] We have guarded against confounding to the extent possible by evaluating for known potential confounders , including ancestry group ( which likely includes HLA ) , BV , age , cohort , ethnic group affiliation and spatially isolated ancestry . The next general step in establishing causal association is replication by different groups and functional studies . We are currently engaged in the latter and encourage validation of our results in genetic association studies with HIV-1 exposure measurement as well as functional studies by others of the variants/genes discovered and replicated here . In summary , we used quantitative measures of HIV-exposure to select individuals with extreme HIV-1 acquisition phenotypes and thereby optimize our power to detect genes associated with risk of HIV-1 acquisition . We identified variants , including rare variants , in CD101 and at least one in UBE2V1 that are significantly associated with increased HIV-1 acquisition risk . More detailed dissection of the molecular basis for this association may identify unique interventions that use these pathways to improve public health prevention of HIV-1 .
We identified individuals for this study from HIV-1 serodiscordant couples recruited into three cohorts of African heterosexual HIV-1 serodiscordant couples: the Partners in Prevention HSV/HIV Transmission Study [20] ( ClinicalTrials . gov number , NCT00194519 ) , the Couples Observational Study [14] , and the Partners PrEP study [22] ( ClinicalTrials . gov number , NCT00557245 ) ( S1 Table ) . Detailed procedures have been reported elsewhere for each of these studies [14 , 20 , 22] . Briefly , routine follow-up visits with both partners were scheduled at least every 3 months , with clinical , behavioral and demographic data collected . HIV-1 seroconversion ( SC ) was assessed by HIV-1 rapid test at the study clinic; positive rapid tests were confirmed by HIV-1 ELISA at the site laboratory , and by Western Blot in batch at the University of Washington ( UW ) . Plasma virus sequencing performed on both partners for each couple associated with SC was used to confirm transmission linkage [66] . All participants provided written informed consent for participation in the clinical study , and samples for this genotyping study were selected from among those participants recruited at 14 sites across all three cohorts who additionally consented to host genetic studies . Relevant study documents went through ethical review and approval by the following committees: An extreme phenotypes case-control design was employed for the Discovery stage . The extreme phenotypes design provides the greatest statistical power for a fixed Discovery sample size of 100 individuals ( with power increasing as the percentiles of phenotype become more extreme in the two arms ) . Hence , individual phenotype was a primary consideration in Discovery stage participant selection . Extreme cases here comprise individuals with relatively low estimated exposure who seroconverted during the study , especially those who converted early in the study . Extreme controls comprised individuals with high estimated risk who remained seronegative over the full observation period with follow-up for at least nine months . Highly-exposed HIV-1 exposed seronegative control individuals with longer follow-up time were considered more extreme based on cumulative exposure scores across all study visits . Cumulative exposure score to rank extremes was calculated as previously described using plasma HIV-1 RNA level of the infected partner , frequency of unprotected sex and male circumcision status [23] , with the modification that all participants must have reported unprotected sex by at least one partner in the couple to be eligible for the Discovery stage sample . It is possible for an individual to have a fairly high exposure score in this model even if no unprotected sex is reported , because the original risk score is based on empirical estimates of risk of seroconversion given the variable values , and some participants who reported no unprotected sex did seroconvert with plasma HIV-1 genomes matching those of their infected partners . Hence , the risk among the group that reported “never unprotected sex” is not zero . Nevertheless , the latter risk is smaller than that for those who report unprotected sex , all else equal , and the risk rises as the proportion of reports with unprotected sex rises ( S6 Fig ) . Seronegative individuals from couples that did not report unprotected sex were excluded for statistical power reasons: including unexposed individuals lowers the statistical power to the point where a p-value of 6 . 3x10-5 in the Replication stage ( N = 261 ) ( Table 3 ) becomes p = 0 . 03 ( N = 1229 ) ( Table 4; replication stage methods below ) . Given these exposure scores/conditions defining the extremeness of phenotype within the potential cases and controls , we then incorporated both gender balance within-group and pairwise ethnicity/sex matching between-groups to this design to avoid happenstance confounding by differing proportions of sex or ancestry in the relatively small samples . The potential impact of ancestry-confounding in this subpopulation was largely unknown when the Discovery stage was designed , and we opted to take this precaution against confounding . Specifically , seroconverter ( SC ) “cases” were selected from the Partners in Prevention HSV/HIV Transmission Study and COS cohorts among couples with laboratory confirmed linked HIV-1 transmission [67] , who were HIV-1 polymerase chain reaction ( PCR ) negative at enrollment and had the lowest exposure scores , conditional on relatively equal numbers of males and females . Individuals from the Partners PrEP cohort ( S1 Table ) were not available for Discovery stage sampling because the trial had not come to complete closure with data available for ancillary studies at that time . For selection of Discovery stage controls , we excluded from consideration: SC individuals ( with either linked or unlinked transmissions ) , couples in which HIV-1 infected partners reported use of any antiretroviral therapy ( ART ) , couples with no reported unprotected sex during the study and couples with less than nine months follow-up time . After these exclusions , for each selected case , all HESN of matching sex and self-reported ethnicity were identified , and from these matched individuals , the HESN individual with the highest cumulative exposure score was selected as the matching control . A total of 65 case-control pairs were identified in this manner , with identification of low exposure ( extreme ) cases being the highest selection priority , followed by ethnicity/sex matching for controls , followed by criteria for high exposure among the matching controls . We quality controlled these case-control pairs for gender check , cryptic relatedness and genetic heterogeneity across multiple longitudinal whole blood DNA samples , using a custom Illumina Goldengate chip with 384 single nucleotide polymorphisms ( SNPs ) . These test SNPs were selected as being the most predictive of ancestry clusters and individual identity from a principal component analysis ( PCA ) on data from a previous genome-wide association study [14] that included samples from these same cohorts . Specifically , we first performed a PCA on a pruned set of 133 , 991 SNPs from that GWAS that had low linkage disequilibrium . The first five PCs for this analysis were effective at distinguishing participants from East and southern Africa , by country ( Kenya , Uganda , Tanzania , South Africa and Botswana ) and by self-reported ethnicities reported in >2% of participants ( S10 Fig ) . Using the first five PCs , we then assigned all participants to one of nine ancestry clusters based on model-based clustering , which has previously been shown to reduce population stratification bias . Subsequently , we used the Random Forests algorithm to identify 357 SNPs that were most predictive of geographic region ( East Africa versus southern Africa ) and the nine ancestry clusters . These 357 SNPs were collectively able to differentiate the ancestry clusters but were much less important for predicting ancestry than self-reported ethnicity and geographic region ( S11 Fig ) . The final Goldengate SNP chip included the 357 ancestry SNPs along with 27 SNPs that maximized the probability that all participants had a different genotype at one or more loci in order to ensure that DNA samples came from unique individuals . The genotyping chip was used on DNA from 65 potential case-control pairs . After eliminating samples that failed QC ( 1 case failed cryptic relatedness , and longitudinal samples from 2 controls suggested potential sample heterogeneity ) , verifying matching on ancestry cluster and identifying controls with highest cumulativeHIV-1 exposure scores over all visits , 50 case-control pairs were selected for complete genome sequencing for the Discovery analysis ( 24 male cases with matched controls , and 26 female cases with matched controls ) . When characterized by the two strongest components of the exposure score , namely mean plasma HIV-1 RNA and proportion of follow-up visits where no-condom use was reported , the Discovery stage controls were verified as sampled from the highest exposure strata ( Table 1 ) . Cases also had a median time to HIV-1 seroconversion of 11 . 8 months , while the median duration of follow-up without seroconversion for controls was 22 . 8 months . WGS was performed by Complete Genomics , Inc ( CGI ) , using published methodology [68] , and with samples blinded as to case and control assignment . SNV calling was performed by CGI using proprietary software cgatools version 1 . 5 . 0 build 31 ( dev ) and human reference genome NCBI Build 37 . Overall , CGI sequence quality and aggregate descriptive characteristics were similar for case and control genomes ( S2 Table ) . Genome annotations were from CGI based on the National Center for Biotechnology Information ( NCBI ) refSeq database . Annotations used for this rare SNV analysis included non-synonymous protein coding sequence , 3’- and 5’-untranslated regions , and splice donor/acceptor sites with exon/intron boundaries identified through the University of California-Santa Cruz ( UCSC ) refFlat database . [69] To validate WGS sequence data we identified 8 SNVs in the WGS data that lacked an rsID in dbSNP and were identified in either of the two genes targeted for replication analysis ( CD101 and UBE2V1 ) . SNVs not previously reported to dbSNP have a higher probability of being false positives than the bulk of SNVs in dbSNP , so these provide “tough” test cases for the CGI calls for validation of true positive rare variants . These SNVs were Sanger sequenced from the individual genomes originally used to generate the WGS . All 8 were successfully validated ( S3 Table ) . Prior to analysis , each SNV was classified as “functional” or not , based on CGI annotation indicating that the SNV either ( 1 ) altered protein coding , ( 2 ) was at a splice-site or ( 3 ) in a UTR sequence region . Only these genic categories of functional polymorphisms were included in this analysis . Case-control comparison of SNVs aggregated by genic region ( “variant burden” ) was accomplished through logistic regression analyses with burden scoring derived specifically for rare variant analysis by Morris and Zeggini with the p-value based on the likelihood ratio test ( their RVT1 ) [24] The Wald test p-value was found to be unreliable—much too conservative—for genic regions with high imbalance between cases and controls , which are exactly the situations we seek to find . Although cases and controls were matched by ethnic origin , the first 3 principal components ( PC ) in the logistic regression of each genic region were used to control for potential residual population stratification . To focus on less common or rare variants and higher effect sizes , the RVT1 ignores variants with MAF above a specified cutoff ( >0 . 125 in our analysis ) . The MAF cut-off of 0 . 125 , which is somewhat higher than typically considered rare , was used to allow for random variation in the observed MAF for rare SNVs in the population , as well as to allow for enrichment of rare SNVs in the sample due to the selection of extremes . Because the study participants are sub-Saharan Africans , external estimates of MAFs for the observed SNVs were not available . This testing prioritization strategy resulted in 18 , 354 genic regions tested ( each region with an RVT1 p-value produced for the null hypothesis of no difference in variant burden within the region between cases and controls ) and included a total of 284 , 632 functional variants in these tests . Principal components created by LD-pruning variants with MAF > 0 . 03 were used to adjust the RVT1 tests . Genic regions with the lowest p-values were considered for replication . These also have the lowest False Discovery Rates [70] ( FDRs ) . Based on FDR analysis , the two most significant regions had a probability of 0 . 83 that at least one was a true positive ( not a false discovery ) . These two regions were moved forward to the MIPs Replication stage based on this high probability of having a true positive result . We planned and implemented a formal frequentist p-value-based criterion for declaring significance of the replication of CD101 and UBE2V1 genic region variants contributing to the aggregate scores in the Discovery stage . We broke the variants into smaller aggregates for the replication analysis ( four aggregates for CD101 , two for UBE2V1 ) in an a priori attempt to further enrich at least some of these aggregates with a higher proportion of true positive variants . This reflected the fact that Discovery stage test regions include both variants that are significantly associated with outcome ( i . e . , signal ) and those not significantly associated with outcome ( i . e . , noise ) . If sub-groups of the originally aggregated variants can be enriched for truly associated variants having a common direction of effect , the statistical power for replication of such an enriched sub-group can be increased considerably relative to the larger , noisier aggregate [19 , 28] despite the need for a multiple-test correction due to increasing the number of tested aggregated variant groups . Based on statistical reasoning , variants with large effect sizes and the same direction of effect are most likely to be driving the Discovery finding . To formulate sub-groups prior to the Replication analysis , “by-variant” association results were tabulated and examined for size and direction of effect . Confidence intervals are wide for these by-variant results due to the low MAF/low-power issue , but they are not completely uninformative . To complement these results , we also compared by-variant results to genome and protein maps of the regions ( Fig 1; S4 & S5 Tables ) . Sixteen of 24 predicted functional variants in CD101 had by-variant estimates with increased risk ( HRs>1 ) , which is the direction of effect for the aggregate RVT1 CD101 result that we seek to replicate . Fourteen of these were designated as PRVs based on significance of by-variant tests ( S3A Fig ) . These 14 CD101 PRVs were divided into Ig-like ( N = 5 PRVs ) , cytoplasmic ( N = 5 ) , 3’-UTR ( N = 2 ) and splice site domain ( N = 2 ) subgroups . Similarly , 11 functional variants in UBE2V1 identified as PRVs ( S3B Fig ) divided into 5’-UTR ( N = 6 ) and 3’-UTR ( N = 5 ) subgroups . Although a fairly common strategy in replication of rare variant results is to include all rare variants within a region in the replication test whether or not these variants were among those identified in the Discovery stage . For example , such a strategy would be used to study low-density lipoprotein receptor ( LDLR ) variants [27 , 28] ( OMIM #606945 ) , given that the structure and function of LDLR is well-known , providing reasonably high confidence that novel missense or truncation RVs in particular sub-regions of the LDLR gene will have a deleterious association . In this study , we elected not to include in replication tests any variants seen only in the Replication stage and not in the Discovery stage for four reasons: ( 1 ) identification of individual variant contributions is important at this point to identify variation in any specific structural protein components affecting the outcome ( and point toward mechanism ) ; ( 2 ) identification of individual variant contributions for variants that are only modestly rare makes it feasible to use genotype tests for presence of individual risk variants in future studies or risk assessments; ( 3 ) addition of new variants in the tests at the Replication stage obscures the contribution of variants from the Discovery stage alone; and ( 4 ) we did not expect a large contribution from new/novel/rare variants in the Replication stage because of the sample size limitations at this point , and therefore were not concerned by potential loss of statistical power via this choice . In addition to the primary replication tests that include subsets of variants identified in the Discovery stage ( four tests planned for CD101 and two for UBE2V1 ) , we performed exploratory testing on all additional variants found in the Replication stage to take the most advantage of these findings while limiting the formal replication tests as above . All exploratory statistical test results are reported by nominal p-value and FDR to provide some quantification of evidence for association , but no formal multiple test correction is given for these findings given their often highly correlated nature ( making a Bonferroni correction misleading ) and due to the post hoc character in some cases . We measured sexual exposure behavior defining Protected-sex Index ( PI ) as the proportion of study visits for which only abstinence or 100% condom use was reported . Simulation studies showed that statistical power was highest when the Replication cohort included individuals with PI ≤0 . 6 while individuals with higher levels of protected sex were excluded . We set the Replication cohort to be the 262 baseline uninfected individuals each having PI ≤ 0 . 6 who remained after the Discovery stage . Due to the economy of sequencing multiple individuals using multiple inversion probe sequencing ( MIPs ) [33] and the intention to demonstrate proof of principle in our power calculations , we successfully sequenced an additional 986 individuals ( thirteen 96-well plates ) . This set of 1248 individuals was balanced on sex and ethnicity relative to seroconverters and non-seroconverters at a 1:6 ratio , a typical sampling strategy . These additional 986 individuals had higher exposure scores than the average participant in the serodiscordant couple cohorts based on high plasma HIV-1 RNA levels of their infected study partner , but many reported low levels of or no unprotected sex . Therefore , this group of additional participants was not considered by itself . There were 19 individuals failing the MIPs procedure due to low DNA concentration , resulting in 1229 individuals with 261 in the Replication sample and 968 in the auxiliary sample . MIPs can be used for targeted sequence capture , followed by massively parallel sequencing of captured products . This strategy is efficient and cost-effective for sequencing multiple candidate genes in modest to large sample sets [33] . MIPs were designed to target all coding exons plus 10 additional base pairs of intron/exon flanking sequence for CD101 ( RefSeq NM_004258 ) , and UBE2V1 ( RefSeq NM_021988 ) . In total , 74 MIPs ( Integrated DNA Technologies , Coralville , IA ) were designed and pooled in equimolar ratios and a test library of 24 African control samples from Centre d’Etude du Polymorphisme Humain ( CEPH ) and 23 Caucasian in-house control samples was produced and evaluated . For MIPs that failed to produce the minimum required average sequence depth of 60X in the test library , their concentration in the final pool was increased according to their level of under-performance in order to increase coverage above the average threshold . The optimized MIP pool ( S12 Table ) was then used to generate libraries and targeted sequence from the ethnically-matched Replication Sample . The MIP pool was phosphorylated with T4 Polynucleotide kinase and T4 Ligase Buffer ( New England Biolabs , Ipswich , MA ) . The reaction was held at 37°C for 45 minutes and then de-activated at 65°C for 20 minutes . For the majority of samples ( N = 1096 ) , 100ng of genomic DNA per sample was used in a capture reaction with a 200:1 MIP to gDNA ratio . For samples with less than 100ng of DNA available ( N = 228 ) , a range of 40ng to 100ng was used in the capture . During a 24-hour reaction at 60°C , HemoKlen Taq ( New England Biolabs , Ipswich , MA ) was used to capture the target regions and Ampligase ( Epicentre , Madison , WI ) was used to circularize constructs . E . coli Exonuclease I and Exonuclease III ( New England Biolabs , Ipswich , MA ) were used to enzymatically clean capture reactions . The reactions were held at 37°C for 30 minutes and then de-activated at 95°C for 2 minutes . Captured products were amplified using iProof HF Master Mix ( Bio-Rad , Hercules , CA ) during 20–22 cycles of PCR . This PCR was performed using a generic forward primer and reverse primers containing a generic portion along with a unique 8bp molecular tag used to barcode the captured products for each sample . Between 40 and 96 barcoded samples were pooled into each library and the libraries were purified using Agencourt AMPure XP ( Beckman Coulter , Indianapolis , IN ) magnetic beads in the ratio 0 . 9:1 beads to samples . Libraries were eluted into EB buffer ( Qiagen , Valencia , CA ) and quantified via the Broad Range Quant-iT dsDNA Assay Kit ( Life Tech , Waltham , MA ) using a SpectraMax Gemini XPS Fluorometer ( Molecular Devices , Sunnyvale , CA ) . Libraries were then pooled and sequenced on an Illumina MiSeq using 300 cycle paired end ( v2 ) reagents ( Illumina , San Diego , CA ) . Custom oligonucleotides complementary to sequences in the MIP constructs were used . Each run contained between 136 and 192 samples . Libraries were diluted and denatured according to Illumina’s standard procedure with final loading concentrations ranging from 8 to 10pM . Individual fastq files were generated by the MiSeq Reporter Software ( v2 . 5 ) . The resulting fastq files were aligned to Hg19 with the Burrows-Wheeler Aligner ( BWA v0 . 5 . 9-r16 ) , and a multi-sample variant call file was generated using the Genome Analysis Tool Kit ( GATK v2 . 4-9-g532efad ) . Annotated was performed using Variant Effect Predictor v82 ( http://www . ensembl . org/info/docs/tools/vep ) . The average sequencing depth per sample across all sites was 198X . A total of 1229 individuals’ sequences passed quality control . Given that a subset of these samples ( N = 138 ) had less than 100ng of total DNA and therefore could have yielded high variant missingness in these samples but with good data quality in the remainder of samples , we set the threshold of 12% missingness for excluding a variant from the Replication stage tests below ( compared to a conventional GWAS threshold of 5% missingness ) . A total of 314 variants passed quality control including a Hardy Weinberg Equilibrium test cut-off of p>3 . 18x10-5 . Testing for association between genomic variation and HIV-1 acquisition risk in the Replication stage was performed using straightforward , standard censored data methods , since follow-up time in the three cohorts was variable with most observations being censored . This provides greater statistical power than using a subset of individuals who have reached a minimum follow-up time or the outcome to create a case-control sample , as it allows all follow-up time to contribute to the estimated risk associations . Kaplan-Meier plots and Cox proportional hazards models were used . We performed the primary replication tests by scoring the aggregate risk variable as 1 , if an individual carried a minor allele for any of the variants in the aggregate , and 0 otherwise . This variable was tested for association with time-to-seroconversion via the Cox model . Risk variables were tested jointly in the Cox model , as well , and results were checked for possible confounding by sex , cohort , age , country/ethnic group ( five countries with Kenya broken into three major ethnic groups plus others , a surrogate for ancestry , S4 Fig ) . Principal components that were available on the subset of Replication individuals from the GWAS also were checked as potential confounders/adjustment variables on this subset . Because we and others [71 , 72] have found bacterial vaginosis ( BV ) to be associated with increased risk of sexually acquired HIV-1 infection , BV is a potential confounder if associated with either CD101 or UBE2V1 . BV assessment ( as we have previously defined it [73] ) was available on a subset of Replication individuals and was tested as a potential confounder on this subset . Cytokine levels at baseline entry to the cohorts were measured in a nested case-control subset of individuals from a previous study [37] , wherein laboratory methods are described in detail . We assessed for differences in these cytokine levels among CD101 variant carriers and non-carriers . This analysis includes all individuals who had both baseline cytokine measurements performed and were among the individuals included in the Replication stage MIPS sequencing or Discovery stage WGS in order to determine their CD101 variant genotypes . Three CD101 missense variants with increased-risk in the Discovery stage data also had high increased-risk in the Replication stage data and were all in the Ig-like aggregate that replicated overall ( chr1:117554421 , chr1:117560058 , chr1:117568500 ) . We separated individuals with cytokine measurements into individuals with any minor allele for any of these three variants and individuals without any of the three alternate variants . Individuals in the second group might have other CD101 variants , but it is impossible to separate individuals with certainty into those with and without any CD101 causal variants; but this grouping will result in a conservative estimate of the difference between groups , as some individuals with causal CD101 variants might be included in the group “without causal” variants and diminish the difference between groups . ( That is , any misclassification here will not produce false positive results but attenuate differences instead . ) We then compared cytokine measurement distributions between these two groups . Since more than half of the cytokines had median values at the limit of detection in the overall group , we compared the two groups for a difference in the 75th percentile rather than the median . This test assesses for a difference in the cytokine distributions primarily using information from those values that are above the limit of detection . ( Five cytokines measured in the original report [37] had 75th percentile values that were at the lower limit of detection and were not included in this analysis ) . Specifically , logistic regression was employed with the outcome being an indicator of whether a cytokine measurement was above the overall 75th percentile for both groups combined . The independent variable was the CD101 carrier status indicator and the regression was adjusted for the panel ( batch ) for each measurement . Several batches had significantly different means for some cytokines , and adjusting for batch generally increased the significance of the difference between gene variant groups . Mean differences in cytokine levels also were compared between groups to ensure that the logistic regression results were consistent with differences in means , though the latter will be affected by the large number of ties at the lower limit of detection . | Antiretroviral drugs for pre-exposure prophylaxis ( PrEP ) or treatment significantly reduce risk of HIV-1 acquisition and transmission , but face challenges of increasing access , maintaining high adherence , and selecting viral resistance . Improved understanding of the molecular determinants of HIV-1 sexual transmission could provide new public health HIV-1 prevention interventions . Factors proven to impact sexual HIV-1 transmission risk include epidemiologic exposure ( e . g . , level of virus in the transmitting partner and frequency of unprotected sex ) , presence of genital inflammation , and host genetic variants common in the population . Rare or intermediate frequency genetic variants are an increasingly recognized reservoir of complex human disease-causing factors , but are not well studied in HIV-1 infection . However , the low frequency of these variants reduces statistical power to detect disease associations . Aggregating variants in a common biological domain ( e . g . , a gene ) can increase power for identifying variants with a common direction of effect . We report comparison of whole genome sequences from HIV-1 exposure extremes—highly-HIV-exposed individuals who remained HIV-uninfected and lower-exposed individuals who became HIV-infected . We discover and replicate associations between HIV-1 risk and aggregate variation in two genes , CD101 and UBE2V1 that increase directly with the level of HIV-1 exposure . These genes may modulate host inflammation thereby identifying molecular mechanisms linking genital inflammation to HIV-1 infection , possibly leading to novel candidate host-directed HIV-1 prevention interventions . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] | [
"blood",
"cells",
"genome-wide",
"association",
"studies",
"innate",
"immune",
"system",
"medicine",
"and",
"health",
"sciences",
"immune",
"physiology",
"cytokines",
"pathology",
"and",
"laboratory",
"medicine",
"body",
"fluids",
"immune",
"cells",
"pathogens",
"immunology",
"microbiology",
"alleles",
"retroviruses",
"viruses",
"immunodeficiency",
"viruses",
"ethnicities",
"developmental",
"biology",
"rna",
"viruses",
"genome",
"analysis",
"molecular",
"development",
"white",
"blood",
"cells",
"genomics",
"animal",
"cells",
"medical",
"microbiology",
"hiv",
"microbial",
"pathogens",
"t",
"cells",
"hiv-1",
"viral",
"replication",
"genetic",
"loci",
"blood",
"plasma",
"immune",
"system",
"people",
"and",
"places",
"blood",
"anatomy",
"cell",
"biology",
"virology",
"viral",
"pathogens",
"physiology",
"genetics",
"biology",
"and",
"life",
"sciences",
"population",
"groupings",
"cellular",
"types",
"computational",
"biology",
"lentivirus",
"regulatory",
"t",
"cells",
"organisms",
"human",
"genetics"
] | 2017 | Whole genome sequencing of extreme phenotypes identifies variants in CD101 and UBE2V1 associated with increased risk of sexually acquired HIV-1 |
The human ABCB1 ( MDR1 ) -encoded multidrug transporter P-glycoprotein ( P-gp ) plays a major role in disposition and efficacy of a broad range of drugs including anticancer agents . ABCB1 polymorphisms could therefore determine interindividual variability in resistance to these drugs . To test this hypothesis we developed a Saccharomyces-based assay for evaluating the functional significance of ABCB1 polymorphisms . The P-gp reference and nine variants carrying amino-acid–altering single nucleotide polymorphisms ( SNPs ) were tested on medium containing daunorubicin , doxorubicin , valinomycin , or actinomycin D , revealing SNPs that increased ( M89T , L662R , R669C , and S1141T ) or decreased ( W1108R ) drug resistance . The R669C allele's highly elevated resistance was compromised when in combination with W1108R . Protein level or subcellular location of each variant did not account for the observed phenotypes . The relative resistance profile of the variants differed with drug substrates . This study established a robust new methodology for identification of function-altering polymorphisms in human multidrug transporter genes , identified polymorphisms affecting P-gp function , and provided a step toward genotype-determined dosing of chemotherapeutics .
Patients vary widely in their drug responses including unpredicted adverse drug reactions that cause a significant loss of lives and a huge toll on health-care costs [1] . Rational selection and dosage optimization of anticancer agents are particularly important due to their narrow therapeutic index and inherent cytotoxicity . Membrane transporters affect drug disposition and response by determining whether or not the level of drug is maintained within the therapeutic index . Of the known human transporters , P-glycoprotein ( P-gp ) is of particular clinical relevance in that this multidrug efflux pump has a broad range of substrates , including structurally and functionally divergent drugs in common clinical use [2–4] . P-Gp belongs to the ATP-binding cassette ( ABC ) superfamily [5] and is encoded by the human ABCB1 gene ( also known as multidrug resistance 1 gene [MDR1] ) . Multidrug resistance caused by ABCB1 amplification is a major obstacle in cancer chemotherapy . In fact , the ABCB1 gene was originally identified because of its amplification in tumor cells that had acquired cross-resistance to multiple cytotoxic anticancer agents [2 , 6–9] . P-Gp is expressed in many tissues , suggestive of a broad physiological role [10 , 11] and functions by pumping cytotoxic drugs and xenotoxins out of cells into the intestinal lumen , bile , and urine , and thus limiting distribution of such compounds to other organs . Genetic heterogeneity of the ABCB1 gene may be a potent determinant of interindividual variability in resistance to multiple drugs including anticancer agents . Furthermore , P-gp can act alone or in combination with other genetic variants , particularly polymorphisms in CYP3A4 , a cytochrome P450 monooxygenase that metabolizes a wide range of drugs [12 , 13] . Naturally occurring null mutations in P-gp have been reported in mice and dogs but not in humans [14 , 15] . Animals carrying a null ABCB1 variant are viable unless challenged by drugs that are substrates for P-gp . Likewise , there may be unidentified human ABCB1 variants that cause a total loss of function . Numerous ABCB1 single nucleotide polymorphisms ( SNPs ) have been identified . However , the correlation of SNPs with ABCB1 expression and P-gp function in clinical pharmacokinetics has been inconclusive . A synonymous 3435C>T SNP has been heavily studied , but its function remains under debate [16] . Moreover , to date there have been no naturally occurring nonsynonymous substitutions with a validated functional consequence [17] . Robust functional assays of P-gp variants at the cellular and molecular levels are needed to address their impact on clinical pharmacokinetics . Since human populations are outbred , and each individual is heterozygous for several million polymorphisms , the impact of ABCB1 variants is difficult to separate from the potential contributions of other variations in an individual . Yeast cells offer an excellent context for functional analysis of foreign eukaryotic transport proteins [18] . Expressing human proteins and their variants in yeast allows the function of individual variants to be assessed directly . The human P-gp can be functionally expressed in the yeast Saccharomyces cerevisiae , where it exports at least some of the same compounds that it exports in human cells [19] . A typical assay for human P-gp function in yeast involves testing its ability to restore growth to cells in the presence of compounds that would otherwise block their growth . This functional complementation in yeast allows the impact of ABCB1 variants found in human populations to be assessed . This study tested the functional consequences of ABCB1 genetic variants found in ethnically diverse populations ( Figure S1 ) [20] . From this dataset ( http://pharmacogenetics . ucsf . edu or http://www . pharmgkb . org ) , we prioritized nonsynonymous SNPs by their predicted impact on P-gp function , selected ten haplotypes carrying high-priority SNP ( s ) , and determined the level of resistance caused by these ABCB1 variants to clinically important drugs . For those variants that altered function , subsequent experiments tested the mechanism of these effects .
As the first step toward functional analysis of the nonsynonymous variants of human P-gp , we tested the sensitivity of yeast strains harboring mutations in major endogenous multidrug transporter genes , PDR5 , SNQ2 , and YOR1 . Combinatorial deletions of these three genes confer sensitivity to a variety of toxic compounds including two anticancer agents , daunorubicin and doxorubicin , which are substrates for human P-gp [21] . The double mutant pdr5 yor1 ( JRY8008 ) displayed increased sensitivity relative to wild-type cells toward doxorubicin , whereas another double mutant pdr5 snq2 ( JRY8004 ) displayed increased sensitivity toward daunorubicin and doxorubicin . The strain that exhibited the greatest drug sensitivity was the pdr5 snq2 yor1 triple deletion mutant ( JRY8012 ) ( Figure 1A ) ( see Table S1 for the strain list ) . This result was reminiscent of bacterial multidrug efflux pumps that produce greater drug resistance in combination than alone [22] . To address the function of human P-gp in yeast , we used a plasmid ( pJR2702 ) that contains a cDNA for the human ABCB1 gene expressed from the promoter for the S . cerevisiae STE6 gene on a multicopy vector [19] . The yeast STE6 gene encodes an ABC transporter that mediates the export of the a-factor pheromone in MATa cells . The cloned cDNA carried the G185V SNP of ABCB1 , and therefore site-directed mutagenesis was used to restore it to the most common allele , referred to as the ABCB1 reference allele in the Pharmacogenetics of Membrane Transporters dataset ( pJR2703 ) ( http://pharmacogenetics . ucsf . edu or http://www . pharmgkb . org ) . Cells expressing the ABCB1 reference cDNA from the multicopy plasmid in the pdr5 snq2 yor1 strain showed highly increased resistance towards daunorubicin and doxorubicin relative to that of the pdr5 snq2 yor1 strain ( Figure 1B ) . Thus the P-gp reference was functionally expressed in these yeast cells . The Pharmacogenetics of Membrane Transporters study identified fourteen nonsynonymous SNPs in 247 healthy individuals from an ethnically diverse population ( Figure S1 ) [20] . These SNPs comprised 25 haplotypes including 15 haplotypes in which the phase relationship of the SNPs was inferred but not directly resolved . SNPs were prioritized for functional analysis by two criteria: the degree of evolutionary conservation [23] and the biochemical severity of the alteration . The extent of evolutionary sequence conservation and thus inferred constraint at a particular residue was observed across ten mammalian species . The severity of missense changes was estimated by the Grantham scale [24] , which formulates the difference in codon substitutions based on chemical dissimilarity of the encoded amino acids . Grantham values range between 5 and 215 , with higher values indicating more radical chemical changes . Out of the 14 nonsynonymous SNPs in the dataset [20] , we chose seven SNPs for functional characterization ( Table 1 ) . We first focused on the five SNPs with highest Grantham values ( >80 ) : M89T , L662R , R669C , A893S , and W1108R . The M89T polymorphic site was not evolutionarily conserved , but the other four sites were highly conserved . In addition , the P1051A SNP was chosen because of its conservation despite a low Grantham value , and the S1141T SNP was included due to its relatively high allele frequency ( 11% in African Americans ) and evolutionary conservation . Although A893S , S1141T , and R669C SNPs are common variants ( minor allele frequency ≥1% in at least one major ethnic group ) , the remaining four chosen variants are observed only once among 494 alleles from different populations . These rare variants ( minor allele frequency <1% ) were included because rare adverse drug reactions may be due to highly penetrant but rare variants . The alignment and allele count of ABCB1 haplotypes based on the 14 nonsynonymous SNPs identified in the previous resequencing project are presented in Table S2 . From the standpoint of functional impact , the R669C SNP was particularly interesting . First , this Arg-to-Cys substitution had the highest Grantham value ( 180 ) among the fourteen SNPs . Second , this SNP was observed twice in the African American population exhibiting a 1% allele frequency , whereas the four chosen rare variants occurred only once . Third , the R669C SNP may be in phase with the W1108R variant . One of the two R669C SNPs was detected in an individual whose ABCB1 gene also contained the W1108R variant , potentially resulting in haplotype R669C-W1108R . This observation prompted us to test whether a R669C-W1108R allele had a unique phenotype relative to alleles carrying each individual SNP . We constructed plasmids expressing P-gp variants by site-directed mutagenesis on the reference plasmid to evaluate the effect of selected SNPs and their combinations on P-gp function . These plasmids ( pJR2703–pJR2712 ) , along with two control vectors ( YEp352 and pJR2713 ) , were transformed into the pdr5 snq2 yor1 strain ( JRY8012 ) . These yeast strains carrying plasmids with ABCB1 variants ( JRY8025–JRY8036 ) were examined for their level of resistance to daunorubicin and doxorubicin on solid medium . Different P-gp variants displayed higher levels of resistance ( A893S-M89T , L662R , and R669C ) or lower levels of resistance ( A893S , S1141T , A893S-R669C , A893S-P1051A , W1108R , and W1108R-R669C ) relative to the P-gp reference ( Figure 2A and 2B ) . The alleles varied widely in their ability to survive on high concentrations of daunorubicin and doxorubicin . The replacement of Arg669 by Cys led to one of the most drastic gain-of-function effects on the ability of P-gp to confer drug resistance . This allele's elevated resistance was compromised when in combination with W1108R . Cells expressing truncated P-gp ( see Materials and Methods ) were indistinguishable from cells transformed with an empty vector with respect to drug resistance . To quantify the extent of drug cytotoxicity in liquid medium , median effective concentration ( EC50 ) values were measured for daunorubicin and doxorubicin for each P-gp variant in liquid culture ( Figure 2C ) . For the majority of the variants , these results were consistent with those observed in the plate assay . However , the plate assay was more sensitive , allowing variants that were indistinguishable from each other in the liquid assay to be ranked . There was a discrepancy between the two drug resistance phenotypes with the A893S and A893S-R669 variants: the variants showed a slightly higher level of drug resistance relative to that of the reference in the liquid assay , but a lower survival in the plate assay . This difference presumably reflects the nature of the two assays: the plate assay measures the level of cell survival on a relatively high fixed concentration of the drug , whereas the liquid assay determines growth rate over multiple drug concentrations . In the plate assay , all variants for daunorubicin and six variants for doxorubicin exhibited statistically significant differences ( p < 0 . 05 ) ( Figure 2B; Table S3 ) . In the liquid assay , three variants for daunorubicin ( A893S-R669C , A893S-M89T , and R669C ) and five variants for doxorubicin ( A893S , S1141T , A893S-M89T , L662R , and R669C ) exhibited statistically significant increases in EC50 values ( p < 0 . 05 ) ( Figure 2C; Table S4 ) . To determine whether the observed differences in drug resistance were due to differences in protein level , we measured the protein level of each P-gp variant by immunoblotting . The mouse anti-P-gp antibody detected P-gps with an apparent molecular mass of 125 kDa , the expected size of unglycosylated P-gp , in membranes from yeast cells transformed with plasmids carrying reference and variant ABCB1 genes , but not in membranes from control cells transformed with an empty vector . The amount of P-gp reference and variants differed by no more than 1 . 5-fold ( Figure 3A ) . The correlation coefficient of the extent of daunorubicin cytotoxicity of each variant relative to the protein level of each variant was 0 . 227 ( Figure 3B ) . Thus the P-gp variants were present at comparable levels and altered drug cytotoxicity in the variants was not due to the differences in protein levels for P-gp . In principle , the differing drug resistance of the variants might reflect differences in their subcellular localization if the SNPs affected the P-gp trafficking . To test this possibility , strains carrying green fluorescent protein ( GFP ) fused in frame to the C terminus of each P-gp variant were evaluated for their subcellular localization patterns . Fluorescence microscopy indicated that the fusion proteins were localized to both the plasma membrane and the vacuolar membrane in living cells ( Figure 3C ) . The localization patterns were growth-phase–dependent: GFP fluorescence was observed mostly in the plasma membrane in mid-log phase cells and became more concentrated in vacuoles when the cells were grown into the stationary phase . The cells carrying each of the GFP-fused P-gp variants fluoresced to similar extents from the same subcellular location under each growth phase . Thus differences in subcellular localization were unlikely to underlie the differences in drug resistance associated with the variants . The relative resistance of each P-gp variant to the structurally similar drugs , daunorubicin and doxorubicin , were quite similar ( Figure 2 ) . Because P-gp can confer cellular resistance to a variety of cytotoxic drugs , we tested whether P-gp variants might exhibit different resistance profiles when tested with additional P-gp substrates , valinomycin and actinomycin D , which are structurally dissimilar from daunorubicin and doxorubicin . Due to the limited solubility of valinomycin in synthetic ( CSM ) –Ura culture medium , determining the EC50 values was not possible . However , determining the EC30 proved sufficient to distinguish among the P-gp variants for valinomycin resistance ( Figure 4A ) . Although some alleles showed similar trends of resistance for valinomycin and daunorubicin/doxorubicin , others ( e . g . , S1141T , W1108R , and W1108R-R669C ) were qualitatively different in their resistances . Yeast MATa ste6 strains , which lack the a factor pheromone transporter , are reported to be more sensitive to actinomycin D than wild-type strains [25] . This prompted us to investigate the interesting possibility that MATa cells are intrinsically more resistant to actinomycin D than MATα cells . Indeed MATα cells were dramatically more sensitive to actinomycin D ( EC50 15 μg/ml ) than MATa cells ( EC50 48 μg/ml ) . To see if the cytotoxicity profile pattern of P-gp variants is changed with actinomycin D , all variants were tested in a MATa ste6 strain ( JRY8572 ) for their levels of resistance to actinomycin D ( JRY8573–JRY8584 ) ( Figure 4A ) . We tested the statistical significance of all comparisons between the reference and each variant for each drug ( Table S4 ) . Five variants ( S1141T , A893S-R669C , A893S-M89T , L662R , and R669C ) exhibited a statistically significant increase in EC50 or EC30 values for two or more drugs . The A893S and A893S-P1051A variants caused an increase in resistance only for doxorubicin and valinomycin , respectively . The compromising effect of W1108R on R669C was obvious in resistance for all four drugs ( Figures 2 and 4A ) . To see if the relative resistance profile of the P-gp variants to one substrate was predictive of the relative resistance profile to other substrates , we determined the correlation coefficient for all combinatorial pairs of the four relative resistance profiles ( Figure 4B ) . The resistance profiles of three anticancer agents ( daunorubicin , doxorubicin , and actinomycin D ) were highly correlated to each other , whereas the resistance profile of valinomycin exhibited a relatively low degree of correlation with those of the other three drugs .
To understand the correlation between ABCB1 polymorphisms and altered cellular pharmacokinetics , we have developed functional assays of P-gp variants in yeast cells . The function of nonsynonymous SNPs was quantitatively measured in isolation from all other variations in the human genome in a yeast-based in vivo assay . The most sensitive measure of drug transport was a colony-counting assay , which provided both qualitative and quantitative measures of drug resistance in yeast expressing reference and variant P-gp . We observed multiple differences caused by the P-gp variants in the level of resistance to the anticancer agents , daunorubicin , doxorubicin , and actinomycin D , and the potassium ionophore valinomycin . The functional consequences of five ABCB1 polymorphisms were previously unknown: the M89T , L662R , R669C , and S1141T variants were associated with increased resistance to two or more drugs; and the W1108R variant strongly mitigated the impact of R669C on gain of P-gp function ( Figures 2 and 4A ) . Due to its high allele frequency ( 11% in African Americans ) , the S1141T SNP in particular deserves further attention to define its clinical significance . As measured by plating efficiency in an acute exposure test , the difference between the reference and most sensitive ( W1108R ) alleles was approximately 30-fold . In a chronic exposure involving growth in the presence of the drug , like most quantitative comparisons of the activity of single amino acid substituted P-gp mutants in the published data , the differences among the P-gp variant alleles in EC50 or EC30 values were modest in most cases . The functional variations can be magnified in clinical practice , especially for anticancer agents due to ABCB1 amplification in cancer patients . In previous studies , the A893 variant , which is the most common SNP , caused either no significant functional impact [20 , 26 , 27] or increased P-gp function for digoxin efflux [28] . The data shown here were able to detect an effect of this allele and uncovered unexpected complexity in the response . In the acute assay , A893S cells were significantly more sensitive to both daunorubicin and doxorubicin than cells with the reference allele ( Figure 2A and 2B ) . In contrast , in the chronic assay the A893S allele was indistinguishable from the reference allele with respect to daunorubicin and slightly more resistant to doxorubicin ( Figure 2C ) . Like variants of facilitated drug influx pumps in the solute-carrier superfamily , P-gp variants that increased function were common . Most random changes in protein sequence are expected to be deleterious or neutral . The significant enhancement of function common to the alleles tested here may reflect a recent adaptation of human populations to local conditions like toxin exposure , leading to selective pressures on medically relevant phenotypes . Interestingly , in Europeans CYP genes encoding drug-metabolizing enzymes show strong signals of very recent positive selection [29] . Despite its distinct chemical structure , the resistance profile of actinomycin D showed a high level of correlation with those of the other anticancer agents , daunorubicin and doxorubicin ( Figure 4B ) . Valinomycin , which lowers the mitochondrial membrane potential , inducing apoptosis in some cell lines [30] , exhibited a low correlation in resistance relative to other drugs , presumably reflecting differences among P-gp variants in recognition or transport of the drugs . The resistance profiles of the S1141T , W1108R , and W1108R-R669C variants showed the largest variation across substrates . Based on this finding , we speculate that the region containing W1108 and S1141 contributes to the substrate discrimination activity of P-gp . To date , all mutations that alter substrate specificity of P-gp have been located in the transmembrane domains [16] . In contrast , all seven SNPs for which functional consequences were determined in this study are located either in the extracellular region ( M89T ) or in the cytoplasmic region ( the remaining six variants ) . We used two widely accepted criteria for predicting the functional effect of uncharacterized SNPs to prioritize for functional characterization ( Table 1 ) . Our data on functional consequences revealed that these predictions were sound: four functional SNPs ( L662R , R669C , W1108R , and S1141T ) scored highly on both criteria , while the two SNPs ( A893S and P1051A ) that showed no significant functional impact had lower scores on evolutionary conservation and chemical dissimilarity , respectively . One exception was the M89T variant that altered function despite being poorly conserved among mammals . Most previous functional studies focused on the impact of individual SNPs rather than that of haplotypes . However , in at least some cases , drug response correlates with the patients' haplotypes rather than individual SNPs [31 , 32] . We tested SNP interactions to see if a compound allele consisting of two SNPs has a unique phenotype different from those of single-SNP alleles . Indeed , it is striking that the strong impact of R669C on P-gp function diminished almost completely when combined with W1108R ( Figures 2 and 4A ) . In contrast , the W1108R variant either alone or with A893S contributed no significant alterations in EC50 or EC30 values . This result highlighted the importance of testing the impact of all substitutions in a gene together and suggests that compensatory SNPs may exist in nature . SNPs in the ABCB1 gene have been implicated in altering drug response or susceptibility to diseases such as Parkinson's disease [33] , inflammatory bowel disease [34] , and renal epithelial tumors [35] . However , in many such cases , the reported effects of ABCB1 polymorphisms are conflicting or inconsistent [26 , 36–38] . This inconsistency may have several causes . First , P-gp expression levels may be modified by nongenetic factors , such as diet and comedications , especially when surgical specimens are studied . Second , previous studies with mammalian cell lines rely on transient expression assays , which swamp the subtle effects of SNPs by variable levels of expression . Third , only a few coding SNPs have been functionally tested , such as A893S and N21D , which our analysis predicted would have a weak functional impact [26] . The use of yeast to evaluate the function of nonsynonymous coding SNPs bypasses these issues and allows the function of single coding SNPs and haplotypes to be assessed directly , independent of all other variations in their original human genome . This “in yeast pharmacogenetics” can function as a robust screening and phenotyping tool to characterize additional SNPs in ABCB1 and presumably other human multidrug transporter genes . During the course of these studies , we observed that MATα cells were highly sensitive to actinomycin D , whereas MATa cells were resistant . This was apparently due to expulsion of the drug by the a cell-specific Ste6 transporter . Perhaps chemical exposures in ecological niches or the consequences of treatment with therapeutics might lead to the extreme mating-type biases observed with some fungal pathogens . For example , the mating-type–specific niches occupied by Cryptococcus neoformans may reflect the ability to transport toxins out of the cell in certain environments [39] .
S . cerevisiae strains used in this study are listed in Table S1 . Standard rich medium ( YPD ) , CSM , and synthetic medium lacking nutritional supplement ( s ) ( CSM–Ura , CSM–His , and CSM–Ura–Trp ) were prepared as described [40] . Yeast cells were grown routinely at 30 °C . A P-gp-expressing plasmid , pJR2702 ( alias pYKM77; a multicopy-number vector ) , was kindly provided by Jeremy Thorner ( University of California , Berkeley , California , United States ) and used for constructing expression plasmids for ABCB1 bearing different SNPs . A cDNA for the human ABCB1 coding sequence ( GenBank accession number M14758 . 1 ) was cloned into a multicopy URA3-marked plasmid with the 2 μm origin of replication ( YEp352 ) and expressed from the yeast STE6 promoter ( pJR2702 ) . Substitutions at the SNP position were carried out in pJR2702 by site-directed mutagenesis with primers designed to generate individual haplotypes ( Table S5 ) , using the QuikChange site-directed mutagenesis kit from Stratagene ( http://www . stratagene . com ) . We introduced five single SNP alleles and four compound alleles consisting of a two-SNP haplotype into the reference plasmid ( pJR2703 ) , creating plasmids pJR2704 to pJR2712 ( Table S1 ) . As a negative control , a −1 frameshift mutation at codon 1 , 200 of the ABCB1 sequence ( 1 , 280 amino acids ) was constructed; this cDNA encodes a truncated product of 1 , 228 amino acids expected to be nonfunctional when expressed ( pJR2713 ) . Presence of the desired substitution in the plasmids was verified by DNA sequencing . These eleven constructs , along with another control lacking the entire ABCB1 sequence ( pJR1016 ) , were transformed into a MATa yeast strain lacking three different ABC transporter genes ( Δpdr5 Δsnq2 Δyor1 , JRY8012 ) , resulting in strains JRY8025 to JRY8036 ( Table S1 ) . Daunorubicin and doxorubicin were kindly provided by Robert Schultz in the Developmental Therapeutics Program of the National Cancer Institute , National Institutes of Health ( NIH ) ( Rockville , Maryland , USA ) . Valinomycin and actinomycin D were from Sigma ( http://www . sigmaaldrich . com ) . For drug cytotoxicity assays , stock solutions of the drug were prepared at 10 mM in 5% DMSO for daunorubicin and doxorubicin , in 98% ethanol for valinomycin , and in 100% DMSO for actinomycin D . In the spotting assay , cultures from each strain were grown to midexponential phase , titrated to the same concentration ( ~107 cells per 1 ml ) , and serially diluted 5-fold . Aliquots ( 4 μl ) from the dilution series were spotted onto a CSM–Ura plate containing the indicated concentration of the drug . Control plates lacking the drug contained the solvent control at the same concentrations as CSM–Ura plates containing the drug . In the plate assay , cultures from each strain were grown to midexponential phase and titrated to the same concentration ( ~105 cells per 1 ml ) . Aliquots ( 100 μl ) were spread onto a CSM–Ura plate containing the indicated concentration of the drug . The same aliquots were further diluted 20-fold ( ~5 , 000 cells per 1 ml ) and spread onto control plates lacking the drug . After incubation for three days , colony numbers per plate were counted . Drug resistance was further assayed quantitatively in 96-well microtiter plates ( Corning , http://www . corning . com ) , containing equal volumes ( 200 μl ) of CSM–Ura liquid medium with different concentrations of the drug . Yeast transformants grown to stationary phase in CSM–Ura were diluted to an OD600 of 0 . 1 . Equal volumes ( 200 μl ) of these diluted cultures containing increasing concentrations of the drug were added to wells and incubated at 30 °C for 24 h in a Tecan microtiter plate reader . Cell growth was monitored in the absence of the drug in the presence of the same solvent as a negative control . For the experiments with liquid medium , the EC50 ( median effective concentration ) value was defined as the drug concentration that reduced growth of the treated cells to 50% of growth of the control cultures as judged by OD600 when the increase in OD600 of the control cultures was 0 . 7 ( midexponential phase ) . To rule out the possibility that variations in copy number affect the observed differences in drug resistance for the vectors bearing each P-gp variant , all measurements were examined in a series of independent transformants for each of the P-gp variants . Membrane fractions of yeast cells with the plasmids bearing ABCB1 variants ( JRY8025–JRY8036 ) were prepared as described [41] . The mouse monoclonal anti-P-gp antibody C219 , kindly provided by Michael Gottesman ( National Cancer Institute , NIH , Bethesda , Maryland , United States ) , was used in immunoblots to quantify the level of P-gp variants in yeast . A rabbit antibody against the Gas1 protein , kindly provided by Randy Schekman ( University of California , Berkeley , California , United States ) , served as a marker of membrane proteins . Human P-gp and yeast Gas1 protein were detected simultaneously on the same blot using infrared-labeled secondary antibodies visualized at two different fluorescence channels , 700 and 800 nm . The blot was developed and quantified by Odyssey Infrared Imaging System ( LI-COR Biosciences , http://www . licor . com ) following the manufacturer's protocol . A codon-optimized GFP gene for yeast , yEGFP1 [42] , was amplified by PCR with oligonucleotide primers designed to allow in-frame fusion to the 3′ end of ABCB1 reference and its variants in a yeast expression vector by recombination following transformation into yeast [43] . The presence of yEGFP in the construct was verified by colony PCR and DNA sequencing . For fluorescence microscopy , cells were grown in synthetic medium without tryptophan to minimize autofluorescence . Imaging was done at room temperature using an Olympus IX-71 microscope equipped with 100× NA1 . 4 objectives and Orca-II camera ( http://www . olympusamerica . com ) . ImageJ ( http://rsb . info . nih . gov/ij ) was used for manipulation of images . The probability of a statistically significant difference between the mean values of two datasets was determined by one-way ANOVA with Dunnett's post-test using GraphPad Prism version 4 . 03 for Windows , GraphPad Software ( http://www . graphpad . com ) .
The Entrez ( http://www . ncbi . nlm . nih . gov/Entrez ) accession numbers for the genes described in this paper are 5243 for human ABCB1 , 1576 for human CYP3A4 , 854324 for yeast PDR5 , 851574 for yeast SNQ2 , 853198 for yeast YOR1 , 853671 for yeast STE6 , and 855355 for yeast GAS1 . The RefSeq ( http://www . ncbi . nlm . nih . gov/entrez/query . fcgi ? db=OMIM ) accession number for human ABCB1 cDNA carried in plasmid pJR2702 is M14758 . 1 . The Online Mendelian Inheritance in Man ( http://www . ncbi . nlm . nih . gov/entrez/query . fcgi ? db=OMIN ) accession numbers are 168600 for Parkinson's disease , 266600 for inflammatory bowel disease , and 144700 for renal epithelial tumors . | Patients often show varied drug responses ranging from lack of therapeutic efficacy to life-threatening adverse drug reactions . Drug therapy would be greatly improved if it were possible to predict individual drug sensitivity and tailor drugs to patients' genetic makeup . Like all other organisms , humans have a set of transporters and enzymes to detoxify and eliminate foreign molecules including drugs . Understanding the function of genetic variants in these proteins is a key goal toward personalized medicine . To that end , we examined the functional consequences of naturally occurring genetic variants in P-glycoprotein , the most versatile human multidrug transporter . A novel method was developed and employed that can identify function-altering variants in human transporters . This methodology was robust and powerful in that the functional effect of genetic variants can be directly assessed in yeast where all confounding variables in humans are excluded . Surprisingly , the majority of single amino acid substitutions were found to cause alterations in resistance to three tested anticancer agents . This study extends the impact of yeast-based medical research to a new niche , pharmacogenomics . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods",
"Supporting",
"Information"
] | [
"biotechnology",
"oncology",
"biochemistry",
"homo",
"(human)",
"genetics",
"and",
"genomics",
"saccharomyces"
] | 2007 | Function-Altering SNPs in the Human Multidrug Transporter Gene ABCB1 Identified Using a Saccharomyces-Based Assay |
Miltefosine ( MF ) is the only oral drug available for treatment of visceral leishmaniasis ( VL ) and post-kala-azar dermal leishmaniasis ( PKDL ) . Although the drug is effective and well tolerated in treatment of VL , the efficacy and safety of MF for longer treatment durations ( >28 days ) in PKDL remains unclear . This study provides an overview of the current knowledge about safety and efficacy of long treatment courses with MF in PKDL , as a strategy in the VL elimination in South Asia . Literature was searched systematically for articles investigating MF treatment in PKDL . A meta-analysis included eight studies ( total 324 PKDL patients ) to estimate the efficacy of MF in longer treatment regimens ( range 6–16 weeks ) . We found a per-protocol ( PP ) initial cure rate of 95 . 2% ( 95%CI 89 . 6–100 . 8 ) and a PP definite cure rate of 90% ( 95%CI 81 . 6–96 . 3 ) . Descriptive analysis showed that 20% of patients experienced adverse events , which mostly had an onset in the first week of treatment and were likely to get more severe after four weeks of treatment . Gastrointestinal ( GI ) side effects such as vomiting , nausea , diarrhoea , and abdominal pain were most common . Longer treatment regimens with MF are effective in PKDL patients in India , however with the caveat that the efficacy has recently been observed to decline . GI side effects are frequent , although mostly mild or moderate . However , on the basis of limited data , we cannot conclude that longer MF treatment regimens are safe . Moreover , VL and PKDL pharmacovigilance studies indicate a risk for serious adverse events , questioning the safety of MF . The provision of safer treatment regimens for PKDL patients are therefore recommended . Until these regimens are identified , it should be considered to halt the use of MF monotherapy for PKDL in order to preserve the drug’s efficacy .
Post-Kala-Azar Dermal Leishmaniasis ( PKDL ) is a dermal complication of visceral leishmaniasis ( VL ) caused by the Leishmania donovani parasite , which is transmitted by phlebotomine sandflies . The PKDL disease can appear weeks to years after the successful cure of VL and is characterised by skin lesions , mainly present on places that are easily exposed to sunlight , such as the face [1] . The prevalence and severity of the disease vary between geographical regions . In East Africa , up to 60% of the former VL patients develop PKDL with mainly maculo-papular skin lesions , which are typically self-healing within three months . In South Asia , only 5–10% of the former VL patients develop PKDL . Most patients have hypopigmented macular lesions , however , up to 20% present with more severe papular or nodular skin lesions . Because spontaneous healing is probably limited [2 , 3] , and may take years , treatment of more severe lesions is indicated . Considering PKDL cases are an important reservoir for transmission , potentially infecting new patients with VL [4] , treatment is also required for public health reasons to achieve control of VL [1] . Because of the high endemicity limited to one geographical region and the availability of good diagnostic and treatment tools , in 2005 The Kala Azar Elimination Program was established as a regional initiative by the governments of Bangladesh , India and Nepal with the goal to eliminate VL in South Asia . Eliminating the PKDL reservoir is an important strategy in VL elimination . The only oral drug available for the treatment of leishmaniasis is miltefosine ( MF , hexadecylphosphocholine ) . This phospholipid derivative was originally developed as an anti-cancer drug but it was found to be unsafe after several studies indicated unacceptable renal- and gastrointestinal toxicity [5 , 6] . Scientists from Germany and the UK discovered the anti-leishmanial effect of the drug in the early 1990s . In 2003 , MF was licensed for the treatment of VL [5] . The drug became the leading compound in the treatment of VL because it was effective , with limited side effects , and oral , so easy to administer [7] . In 2011 , MF was added to the list of Essential Medicines by the WHO . A substantial number of studies evaluated the safety and efficacy of MF in standard VL treatment of 28 days . Clinical trials have mainly been conducted in India , specifically in the state of Bihar , where VL is endemic [8] . Cure rates in VL patients range between 90–100% in a regular dose of 2 . 5 mg/kg per day for children aged 2–11 years; for people aged >12 years and < 25 kg body weight , 50 mg/day; 25–50 kg body weight , 100 mg/day; > 50 kg body weight , 150 mg/day; orally for 28 days . The safety concerns regarding MF mainly relate to its effect on the gastrointestinal tract [8] . Frequently observed adverse events in MF treatment regarding gastrointestinal toxicity that led to treatment interruption are nausea , vomiting , loss of appetite and diarrhoea . Other commonly observed toxicities are related to liver- and renal functions ( e . g . elevated creatinine and ALT and AST levels ) . However , these are often not clinically relevant and normally stabilize during treatment [8] . In addition , animal studies have showed teratogenicity and impaired fertility in men and women , meaning that the use of MF could negatively influence the fetal congenital development . Impaired male fertility in humans as a consequence of miltefosine is currently under assessment by the FDA . Miltefosine was first used in treatment of PKDL in 2006 [9] . In comparison to VL , PKDL requires longer treatment durations with MF . The drug is currently used as first-line treatment for at least twelve weeks in PKDL infected patients in India , Nepal , and Bangladesh [10] . PKDL requires longer treatment durations because of the limited skin penetration of antileishmanial drugs , and the fact that there is no other clinical marker for cure than disappearance of lesions , which may take more than one year in case of macular lesions [1] . Only few studies have investigated the safety and efficacy of the long-term MF treatment for PKDL and sample sizes in those studies are relatively small . Due to the slow clearance of MF in the body concerns are raised regarding the safety and efficacy of the drug in long-term treatment for PKDL . Therefore , this study aims to provide an overview of the current knowledge about safety and efficacy of longer treatment regimens ( >28 days ) with MF in PKDL patients , in order to contribute to the control of leishmaniasis .
This was a systematic review including a quantitative meta-analysis of data from different studies , in order to provide more accurate estimates of the effects of MF treatment in PKDL patients . This study was carried out in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses ( PRISMA ) statement [11] . The databases PubMed and Cochrane library were searched systematically using the following search terms: Miltefosine or hexadecylphosphocholine , Post-kala-azar dermal leishmaniasis , visceral leishmaniasis , kala-azar , safety , efficacy , tolerability , toxicity , clinical effectiveness , adverse events and South-Asia , India , Nepal , Bangladesh ( Table 1 ) . The total number of hits was 146 . Fig 1 shows the flow diagram of the literature search . In addition to the computer search , reference search of all reviewed articles was performed to identify articles missed through the database search . One article was found manually . Inclusion criteria were miltefosine monotherapy , VL or PKDL , human study population , and articles had to be written in English . Articles were excluded based on geographical location ( America , Europe and Africa were excluded ) , in case the study used MF for treatment courses of 28 days or less and in case the study was conducted on animals . There were no further restrictions on age , sex or publication date . All included articles were assessed on basic characteristics such as aim , methodological approach , sample size , treatment dose , treatment duration , conclusions and scientific quality . Primary outcomes of the current review were efficacy and safety . Efficacy was expressed in per-protocol ( PP ) cure rates and Intention-to-treat ( ITT ) cure rates at the end of treatment ( i . e . initial cure rate ) and at the end of follow up ( i . e . definite cure rate ) . Safety was displayed in adverse events and abnormal haematological parameters during or after treatment with MF . The seriousness of these toxicities was rated according to the Common Terminology Criteria for Adverse Events ( CTC ) of the National Cancer Institute [12] . Grades ranged from 1 to 5 ( mild , moderate , severe , life-threatening and death ) . In case of mild and moderate severity ( CTC grade 1 and 2 , respectively ) , patients had to be treated with additional medication . In case of severe and life-threatening severity ( CTC grade 3 and 4 , respectively ) , treatment with MF had to be discontinued . Data management and analyses were performed using SPSS version 25 . 0 [13] . Pooled estimates of initial and definite PP cure rates were calculated by random-effects regression analysis , using Wilson’s Macros for meta-analysis ( Wilson , version 2005 . 05 . 23 ) , after applying sample weights according to sample size . Moderator ( subgroup ) analysis was performed to indicate estimated cure rates for different duration treatment groups ( a dummy variable was created for 6 , 8 , 12 and 16 weeks of treatment ) . Heterogeneity between studies was assessed using Cochran’s Q included in the meta-analysis function . A p-value of <0 . 05 indicated significant heterogeneity .
The cure rates at the end of follow up ( definite cure ) per study and the results of the meta-analysis are displayed in Fig 2 . Meta-analysis showed an estimated PP definite cure rate of 90 . 0% ( 95%CI 81 . 6–96 . 3 ) and the average ITT cure rate was 74 . 9% . The lowest PP and ITT definite cure rates , 57% and 55% , respectively , were found in the study of Ghosh et al [17] . These numbers are substantially lower than the definite cure rates found in the other studies , which can be explained by the high number of treatment discontinuations due to severe side-effects in this study . In several study-arms , all patients , in at least one trial arm , were cured at 12-month follow up [14 , 17 , 19 , 20] . As can be seen in Table 2 , the ITT definite cure rates ranged from 43–100% . The low ITT cure rates in the studies of Moulik et al [14] , Ramesh et al [15] and Ghosh et al [17] ( 45% , 43% and 55% , respectively ) , are strongly influenced by high lost-to-follow-up in those studies . In the study of Moulik and colleagues [14] , the drop-out-rate was no less than 57% . In the 8-week study arm of Ramesh et al [15] and in the study of Ghosh et al [17] the drop-out-rates were 33% and 35% , respectively . The cure rates at the end of treatment ( initial cure ) per study and the results of the meta-analysis are presented in Fig 3 . Five of the eight included studies reported an initial cure rate [15 , 18–21] . Meta-analysis showed an estimated per protocol initial cure rate of 95 . 2 ( 95%CI 89 . 6–100 . 8 ) . As can be seen in Figs 2 and 3 , there seem to be outliers regarding both the initial and definite cure rates ( i . e . numbers that lay outside of the 95%CI of the pooled estimates ) , which indicates heterogeneity . Analysis indicated the degree of variance in and between studies . In the analysis for initial cure rate , significant heterogeneity was indicated ( Q = 15 . 6 , I2 = 61 . 6% and P<0 , 05 ) . 61 . 6% of the variance can be contributed to true heterogeneity . In the analysis for definite cure rate , no significant heterogeneity was indicated ( Q = 13 . 4 , I2 = 25 . 1 and P>0 , 05 ) . 25 . 1% of the variance can be contributed to true heterogeneity . In addition to the estimated overall cure rates , subgroup meta-analysis was performed to indicate the estimated cure rates per treatment group related to treatment duration . Table 3 shows the outcomes of this analysis with treatment duration as moderator variable . No significant differences were found in initial and definite cure rates between the different treatment durations . Studies that were conducted in the past five years show a lower average cure rate ( 92 . 6% and 85 . 7% for initial and definite cure , respectively ) than studies that were conducted more than five years ago ( 98 . 7% and 97 . 7% for initial and definite cure , respectively ) . However , these differences were not statistically significant ( p = 0 . 142 and p = 0 . 081 for initial and definite cure , respectively ) . Nearly 20% ( n = 64 ) of all patients experienced adverse events . The most common side effects reported in the included studies are related to gastrointestinal ( GI ) adverse events . GI side-effects reported were nausea , vomiting , abdominal pain , diarrhoea or a combination of these events . All included studies reported that vomiting occurred in the majority of their patients . Vomiting was graded CTC 1 or 2 in nearly 10% of all patients ( n = 20 ) , however data was lacking in most studies regarding those mild and moderate adverse events . Vomiting with CTC grade 3–4 was experienced by three patients . In addition to vomiting , abdominal pain was reported in three studies ( n = 10 patients ) and graded CTC 1–3 . In patients that experienced events graded CTC 3 or 4 , treatment was discontinued . Events graded CTC 1 or 2 were treated symptomatically . In one study , six patients were treated with additional medication ( Ondansetron ) prior to taking MF in order to reduce repeated vomiting ( CTC grade 2 ) [15] . In one study [17] , treatment was reduced to twelve weeks because of unacceptable side effects . Besides observable side effects , six studies provided data on haematological and laboratory tests performed before , during and after treatment . Laboratory abnormalities were seen in liver function ( elevated bilirubin , SGOT and SGPT ) and kidney function ( elevated creatinine and serum asparate aminotransferase ) . However , in all but one patient , all of these laboratory abnormalities were non-severe and stabilised during treatment without interventions ( e . g . additional medication , or treatment interruption ) . In one patient , an elevated bilirubin sample was graded CTC 2 [18] . In addition to the above-mentioned adverse events , one patient suffered from a cerebrovascular accident ( CVA ) [17] . This serious neurological condition ( CTC grade 4 ) had most likely occurred as a result of the treatment with MF [17] . Ghosh et al [17] investigated the causality association between MF and the CVA with the ‘Naranjo adverse drug reaction probability scale’ [17] . However , an explanation for this association was not provided in the article . The data provided about the time of onset of MF side-effects was lacking in the included articles . The studies of Ramesh et al [15] and Sundar et al [16] did not mention at what time during or after treatment the reported adverse events had occurred . In two studies was mentioned that the GI side-effects occurred during the first weeks of treatment . A few studies provided more concrete data on the days , or weeks , of onset of adverse events . In one study , unacceptable GI side-effects started after four weeks of treatment [15] . In addition , one study provided information on the day of onset for all gastrointestinal side effects [18] . The days of onset for vomiting graded CTC1 were: 32 , 33 , 38 , 39 , 48 , 52 and 69 , and vomiting graded CTC2 were: 33 and 77 . Overall , adverse events were likely to occur in the first week of treatment , but became more severe after six weeks .
The strength of this study is the meta-analytic design . Literature on the safety and efficacy of long-term treatment with MF is scarce and sample sizes are small . Therefore , combining the existing studies in a meta-analysis provides a more accurate estimate of the cure rates in endemic populations in South Asia . However , the results of this review need to be seen in the light of some limitations . First , all included studies were conducted in India , mainly in the state of Bihar . Although the majority of patients treated with long-term MF are Indian patients , one should be careful to generalise the results of this study to other endemic countries in South Asia . Secondly , the meta-analysis showed significant heterogeneity between studies , indicating that the variation in and between the studies was not based on standard error alone but can be contributed to methodological variations between studies ( e . g . different assessments of cure , inpatient versus outpatient , and different research designs ) . Thirdly , the results of later studies may be affected by a decreased susceptibility to miltefosine and the overall efficacy we found may no longer reflect the reality on the ground . In order to eliminate kala-azar in South Asia , PKDL patients need to be treated effectively . This review showed that treatment regimens with MF of six weeks or longer are effective ( up to 90% ) in PKDL patients in India , however with the caveat that the efficacy has recently been observed to decline . There is no straightforward answer to whether MF is an appropriate choice for the treatment of PKDL . This review showed that GI side effects are frequent in longer MF treatments , although mostly limited to mild or moderate side effects . However , on the basis of limited data included in this review , we cannot conclude that longer MF treatment regimens are safe . Moreover , information from previous VL studies and PKDL pharmacovigilance indicate a risk for serious , irreversible or even fatale adverse events , questioning the safety of longer treatment regimens with MF . The highly common GI side effects can lead to non-compliance and form a risk for drug resistance . For this reason , directly observed treatment where possible , adequate surveillance of MF susceptibility in both PKDL and VL patients , as well as drug sensitivity monitoring in parasite isolates is required . The provision of other treatment regimen for PKDL patients are highly recommended . It may be put under consideration to halt the use of miltefosine monotherapy for PKDL and proceed with safer alternative regimen . This will also help preserve the drug’s efficacy . In parallel , research into new treatment regimens should be encouraged . | In this study , we reviewed the available literature on the subject of safety and efficacy of the oral drug miltefosine in the treatment of post-kala-azar dermal leishmaniasis ( PKDL ) . Literature was searched systematically in the PubMed database and eight articles , with a total of 324 PKDL patients , were included . A meta-analysis was performed to estimate the percentage of patients cured after longer ( >4 weeks ) miltefosine treatment . An estimated 90% of patients was found to be cured one year after treatment with miltefosine . In addition , descriptive analysis showed that nearly 20% of the PKDL patients suffered from side-effects . The majority of these side-effects , such as vomiting , nausea , diarrhea and abdominal pain , were mild and related to the gastro-intestinal tract . The findings of this study show that miltefosine is effective , although the efficacy has been observed to decline . The gastro-intestinal side effects were frequent but mostly mild . However , based on the limited data in this study we cannot conclude that longer treatment regimens with miltefosine are safe . In order to preserve the drug’s efficacy , we suggest it may be put under consideration to halt the use of miltefosine monotherapy for PKDL until alternative treatment regiments ( e . g . short combination therapies including miltefosine ) are identified . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [
"medicine",
"and",
"health",
"sciences",
"pathology",
"and",
"laboratory",
"medicine",
"clinical",
"research",
"design",
"vomiting",
"statistics",
"metaanalysis",
"tropical",
"diseases",
"database",
"searching",
"parasitic",
"diseases",
"toxicology",
"research",
"design",
"physiological",
"processes",
"toxicity",
"mathematics",
"signs",
"and",
"symptoms",
"pharmaceutics",
"neglected",
"tropical",
"diseases",
"pharmacology",
"research",
"and",
"analysis",
"methods",
"infectious",
"diseases",
"zoonoses",
"mathematical",
"and",
"statistical",
"techniques",
"adverse",
"reactions",
"adverse",
"events",
"protozoan",
"infections",
"diagnostic",
"medicine",
"physiology",
"leishmaniasis",
"database",
"and",
"informatics",
"methods",
"biology",
"and",
"life",
"sciences",
"physical",
"sciences",
"drug",
"therapy",
"statistical",
"methods"
] | 2019 | The safety and efficacy of miltefosine in the long-term treatment of post-kala-azar dermal leishmaniasis in South Asia – A review and meta-analysis |
Capturing nature’s statistical structure in behavioral responses is at the core of the ability to function adaptively in the environment . Bayesian statistical inference describes how sensory and prior information can be combined optimally to guide behavior . An outstanding open question of how neural coding supports Bayesian inference includes how sensory cues are optimally integrated over time . Here we address what neural response properties allow a neural system to perform Bayesian prediction , i . e . , predicting where a source will be in the near future given sensory information and prior assumptions . The work here shows that the population vector decoder will perform Bayesian prediction when the receptive fields of the neurons encode the target dynamics with shifting receptive fields . We test the model using the system that underlies sound localization in barn owls . Neurons in the owl’s midbrain show shifting receptive fields for moving sources that are consistent with the predictions of the model . We predict that neural populations can be specialized to represent the statistics of dynamic stimuli to allow for a vector read-out of Bayes-optimal predictions .
Predicting the future position of an object in the environment is a common and critical component of many tasks that involve reaching or orienting toward moving targets [1–4] . To execute these prediction tasks successfully , motor plans must extrapolate beyond accumulated sensory input to account for delays in sensory and motor processing , as well as for the future movements of the object . The ability to make accurate predictions of the location of a moving target is especially critical in prey capture . Prey capture for moving targets has been studied at the behavioral and neural levels for animals that rely on visual [5–9] and auditory [10–12] information . For example , salamanders use visual input to predict the position of moving prey , make a head orienting movement toward the target , and then generate a ballistic movement of the tongue to capture the prey [7] . Barn owls also visually track their prey when possible [13] , but are additionally able to use auditory information to capture moving prey [10] . After estimating a sound source’s trajectory , the owl makes a head orienting movement to localize a moving target before preparing to bring its feet forward to strike the prey [10] . Interestingly , salamanders and barn owls have neurons with similar specialized receptive fields that shift in time to mediate predictive prey capture [12 , 6 , 8] . These specializations occur in the fast-OFF retinal ganglion cells of the salamander [6 , 8] and the auditory spatially selective neurons in the optic tectum ( OT ) of the barn owl [12] . The receptive fields of these neurons shift toward a moving source , where the amount of shift is sufficient to account for delays in sensory and motor processing . Furthermore , it has been shown in the salamander that it is possible to read out the predicted location of a moving target from the fast-OFF retinal ganglion cells using a population vector average ( PV ) [8] . Here , we use the PV to model the computations performed by the barn owl as it tracks a moving sound source and address how such a neural circuit may approach optimal performance . These studies open several questions about the neural basis of predictive behaviors . What information is represented in these populations of neurons ? Is the observed neural representation an optimal solution to the prey capture problem faced by each species ? An optimal solution to the prediction problem would take into account the source dynamics , sensory statistics , and prior information to guide the solution . This approach to an optimal solution can be formulated as Bayesian prediction [14] . There is support for Bayesian models of perception and behavior in diverse tasks across multiple species [15–18] . Additionally , there have been multiple proposals for how neural systems can implement Bayesian inference [19–23 , 16 , 24–26] . In particular , several studies have addressed the problem of inference in time in the context of hidden Markov models [20 , 27 , 28] and tracking using the Kalman filter [29 , 19 , 22 , 30 , 31] . However , it remains unknown how a neural system can perform Bayesian prediction . Here we specify how a population of neurons should respond to a moving stimulus to allow for a Bayesian prediction to be decoded from the neural responses . We approach this question in the context of auditory-based prey capture by the barn owl . The Bayesian prediction problem we consider is that of predicting a sound source’s future direction , given a sequence of sensory observations and a prior distribution for direction and angular velocity . It has been shown that the owl’s sound localization for brief sounds is consistent with a Bayesian model [24] . Here , evolutionary pressure for optimality may be expected , given the dependence of owls on successful sound localization during hunting . The success of the PV in decoding predictive movements of visual targets in the salamander [8] and dragonfly [32] makes this a viable candidate mechanism for implementing Bayesian prediction . It has been shown that the owl’s map of auditory space decoded by a PV is consistent with the owl’s localization behavior for brief stationary stimuli [24] . More generally , it has been shown that a population code can encode the statistical properties of the environment to allow a PV to match a Bayesian estimator [23 , 26 , 24] . This model for the neural implementation of Bayesian inference is attractive because it matches the common observation that population codes are adapted to natural statistics [33] . However , the applicability of the PV model to Bayesian inference in time is unknown . Here , we determine the conditions under which a population of neurons with spatial-temporal receptive fields can perform Bayesian prediction for moving sound sources .
We consider the problem of predicting the future direction of a moving source from a temporal sequence of auditory observations . Specifically , the prey capture problem is that of predicting the direction of a moving sound source a short time in the future based on the sequence of interaural time difference ( ITD ) measurements from the sounds reaching the left and right ears ( Fig 1 ) . ITD is the difference in the arrival time of sounds at the two ears and is a primary cue for localization in the horizontal dimension [34 , 35] . The Bayesian filtering approach to predicting at time k the direction at a point n time steps in the future θk+n , given the sequence of observations up to time k , ITD1:k = [ITD1 , ITD2 , … ITDk] , is to compute an estimate from the posterior distribution pk+n ( θ , ω|ITD1:k ) . The form of the posterior distribution is determined by a model for the dynamics of the moving target and the statistical relationship between the state of the target and the ITD observations . The temporal dynamics of the horizontal direction of the moving target are modeled as θk=θk−1+Δtωk−1+ηk ωk=ωk−1+νk where θk is the target direction , ωk is the angular velocity , ηk is a zero-mean circular Gaussian noise process , νk is a zero-mean Gaussian noise process , k is the current time step , and Δt is the time step duration . The sensory information ITD is modeled as a sinusoidal function of direction that is corrupted by noise: ITDk=Asin ( 2πfθk ) +ξk where ξk is a zero-mean Gaussian noise process with standard deviation 12 . 5 μs and the amplitude A and frequency f are determined by the shape of the owl’s head and facial ruff [24] . The sinusoidal mapping between direction and ITD is based on direct measurements of ITD for the barn owl [36] . All noise processes are assumed to be mutually independent and uncorrelated across time . The noise process νk influencing the prey velocity depends on the type of behavior displayed by the prey . A large standard deviation of the noise corresponds to irregular fleeing behavior displayed by prey under close attack when there is no place to hide [37] . A small standard deviation produces a smoother trajectory for the prey , which corresponds to escape toward cover [37] . Here we use a velocity-noise standard deviation of 0 . 125 deg/s corresponding to mouse escape behavior under close-distance owl attack where prey trajectories are smooth [37] . This parameter value has the effect of keeping the velocity roughly constant over a short period of time . The prior depends on both the natural prey behavior and the owl’s bias as determined by the behavioral cost function [24] . Here we assume that the prior emphasizes directions at the center of gaze [24] and slow source velocities . We also assume that there is a weak negative correlation between direction and velocity such that there is a bias for sources moving into the center of gaze [38 , 39] . The form of the prior is a Gaussian with zero mean for both direction and velocity . The standard deviation for direction is 23 . 3 deg [24] , the standard deviation for velocity is 50 deg/s , and the correlation between direction and velocity is -0 . 05 . The parameter values for the velocity standard deviation in the prior ( σv0=50 deg/s ) and during movement ( σvk=0 . 125 deg/s , k ≥ 1 ) describe a situation where the initial velocity can take on a wide range of values , but the velocity will be roughly constant over a short period of time . The Bayesian prediction at time k of the direction at a point n steps in the future , θk+n given the sequence of observations ITD1:k is computed as the mean of the posterior n steps in the future pk+n ( θ , ω|ITD1:k ) Because we are estimating a circular variable , the Bayesian prediction is the direction of the Bayesian prediction vector , defined as the vector that points in the direction of the mean value of the direction n steps in the future: BVk=∫u ( θ ) pk+n ( θ|ITD1:k ) dθ , where u ( θ ) is a unit vector pointing in direction θ ( Methods ) . Solving the Bayesian prediction equations may be computationally difficult for nonlinear or non-Gaussian models [40] . If the system is linear with Gaussian noise , then the Kalman filter can be used for Bayesian prediction [41] . Our model includes Gaussian noise but the mapping from direction to ITD is nonlinear . We found that relationship between direction and ITD is nearly linear for sound sources in the frontal hemisphere ( Fig 2 ) . The root-mean-square ( RMS ) error between the measured ITD and the linear approximation ITD = 2 . 67 μs / deg × θ was 15 . 1 μs for directions between -100 deg and 100 deg . We therefore used the Kalman filter to perform Bayesian prediction for computational simplicity ( Methods ) . The Bayesian model successfully predicts future directions of the prey for smoothly moving sources ( Fig 3 ) . We chose the prediction time step n in order to predict the source direction 100 ms in the future [12] . Initially the Bayesian prediction is dominated by the prior distribution , which emphasizes central directions ( Fig 3A ) . Because of the influence of the prior , the posterior does not initially lead the source direction . However , after a short delay the posterior pk+n ( θ , ω|ITD1:k ) predicts the future direction of the source ( Fig 3A and 3B ) . Note that the performance of the Bayesian prediction differs from the Bayesian tracking estimate . Whereas the tracking algorithm seeks to place the center of posterior at the current source position ( Fig 3C ) , the prediction algorithm seeks to place the center of posterior at the future position of the source . Also , the predictive posterior ( Fig 3A ) is wider than the posterior for tracking ( Fig 3C ) because uncertainty increases as the time window for prediction increases beyond the current time where observations are available . It has been shown that the owl’s map of auditory space decoded by a PV is consistent with the owl’s localization behavior for brief stationary sounds [24] . Here we investigate conditions on a population of neurons with spatial-temporal receptive fields under which the PV will match the Bayesian prediction in time . The PV at time k is given by an average of weighted preferred direction vectors: PVk=1N∑j=1Na ( ITD1:k|θ ( j ) , ω ( j ) ) u ( θ ( j ) ) where the preferred directions θ ( j ) are defined by the motor output . The PV at time k depends on the sequence of past ITD measurements and predicts the future direction of the target . By associating each neuron with a fixed preferred direction θ ( j ) , we are making the assumption that the motor neurons that the OT neurons ultimately influence are fixed . This assumption means that the effect of a given level of response for an OT neuron on the motor output stays constant . The rate function a ( ITD1:k|θ ( j ) , ω ( j ) ) is the firing rate of the jth neuron in response to the sequence of ITD values ITD1:k . We now state our main result , which specifies sufficient conditions so that the PV will approximate the Bayesian prediction estimate . The first prediction derived from our result is that neurons implementing Bayesian prediction using this type of population code will have receptive fields that shift in time towards the moving source ( Fig 4A–4D ) . This is the type of shift that is necessary to compensate for delays and allow for the owl to capture the moving source [6 , 12 , 8] . These delays include signal processing in the brain as well as motor delays , and total approximately 100 ms [12] . While the receptive fields shift in time , there is a delay to the onset of the shift of the receptive field . This delay in the shift occurs in the Bayesian model because the response is initially dominated by the prior before sufficient sensory information has been accumulated . Therefore , the predictive posterior initially lags behind the source direction ( Fig 3A ) . It is only after a delay that the predictive posterior leads the current source direction . The model also predicts that receptive fields get sharper with time . The sharpening of the receptive fields follows the sharpening of the posterior as more sensory information is collected ( Fig 3A ) . Additionally , the model predicts that the shift of the receptive field depends on the speed of the moving source . Faster source velocities lead to larger shifts , while slower source velocities correspond to smaller shifts of the receptive field ( Fig 5A ) . This prediction follows from the fact that the posterior shifts faster for faster sources . The receptive field shifts predicted by the model are consistent with experimental results in the barn owl [12] ( Figs 4 and 5 ) . Neurons in the owl’s OT that are involved in generating head orienting movements show shifting receptive fields for moving sources [12] ( Fig 4E and 4F ) . The receptive field shifts in the owl are consistent with the Bayesian prediction model in that the shift toward the source is not instantaneous , but occurs after a delay ( Fig 4E and 4F ) . Receptive fields of midbrain neurons also get sharper in time , as predicted by the model [12 , 43] . Additionally , the size of the shift varies with the speed of the moving source ( Fig 5B ) . The time course and magnitude of the observed shifts correspond well to the predicted shifts in the model . The model predicts an asymmetry in the shifts of the receptive fields for sounds moving into and out of the center of gaze that increases with the eccentricity of the receptive field ( Fig 6 ) . For neurons with receptive fields at the center of gaze , the shifts for clockwise and counterclockwise sources are mirror images ( Fig 6A–6C ) . For neurons with more peripheral receptive fields , the shifts for clockwise and counterclockwise moving sources are asymmetric ( Fig 6D–6I ) . For neurons with peripheral receptive fields , the initial shift of the receptive field for sources moving into the center of gaze is in the opposite direction than one would expect based on the idea that receptive fields should move towards the source . This occurs because of the effect of the prior on the performance of the posterior ( Fig 3A ) . Initially , the posterior is dominated by the prior and thus at stimulus onset is not leading the source by the desired 100 ms . The asymmetry of the receptive field shifts for peripheral OT neurons has not been investigated in the owl . However , neurons in the owl’s external nucleus of the inferior colliculus ( ICx ) do have an asymmetry in their direction selectivity for sounds moving into and out of the center of gaze , which may be related to asymmetric shifts [38 , 39] . Testing this prediction will require further study . The prediction of asymmetry in the receptive field shift for clockwise moving and counterclockwise moving sources depends on the presence of a prior that emphasizes central directions . We found that predicted receptive field shifts were symmetric for clockwise moving and counterclockwise moving sources in both central and peripheral neurons when the prior in the model was uniform ( Fig 7 ) . As noted above , the asymmetry is caused by the initial dominance of the prior on the location of the peak in the posterior . When the prior is uniform , this effect is removed and the posterior can quickly lead the source direction for motions both into and out of the center of gaze . The receptive field shifts predicted by the model were robust to parameter variation ( Fig 8 ) . We examined the receptive field shifts for different standard deviations of the noise terms and different prior standard deviations for direction and velocity . The model predicted similar magnitudes of shifts for the chosen values ( center column ) and when each parameter was halved ( left column ) or doubled ( right column ) . Changing the standard deviation of the noise corrupting ITD had the greatest effect on the receptive fields ( Fig 8A–8C ) . This parameter influences the width of the posterior and therefore influences the width of the receptive field . The net effect of the receptive field shifts is that the activity moves across the population so that it predicts the future direction of the moving source ( Fig 9A ) . It is this activity that must be decoded by the PV to approximate the Bayesian prediction . To test the PV implementation of Bayesian prediction , we constructed a model of 5000 Poisson neurons with receptive fields that shift according to the posterior ( Methods ) . The PV matched the Bayesian prediction closely for different stimulus conditions ( Fig 9 ) . The PV approximated the Bayesian prediction to within 3 degrees ( root-mean-square ( RMS ) error ) for velocities up to 125 deg/s ( Fig 9B ) . The RMS error in the approximation of the Bayesian prediction by the PV depended strongly on the fraction of time the predicted source direction was in the frontal hemisphere ( spearman rank correlation = 0 . 92; Fig 9C ) . Since all of the preferred directions of the model neurons are in the frontal hemisphere , the model will necessarily fail when the posterior is localized at source directions behind the head . We also computed the RMS error using a population of deterministic neurons to determine the contribution of the Poisson variability of the neurons to the error ( Fig 9D ) . The Poisson variability increased the RMS error for many trajectories ( mean ± s . d . ratio of RMS error for deterministic neurons to RMS error for Poisson neurons 0 . 43 ± 0 . 23 ) . However , the largest errors in the approximation are primarily due to the limited range of preferred directions of neurons in the population . The pattern of error as the initial direction and velocity of a moving source varied is explained by larger errors occurring when the predicted source trajectory spends more time behind the head .
We showed that the PV can read out the Bayesian prediction in time from a population of neurons . The PV will approximate the Bayesian prediction when the population has specialized responses with shifting receptive fields . The types of shifting receptive fields predicted by our analysis are observed in the OT of the owl [12] and the retina of the rabbit [6] and salamander [6 , 8] . This result shows that with the appropriate encoding of the stimulus , a simple decoding algorithm can perform complex computations [44 , 19 , 8] . Our work provides a theoretical framework in which to interpret observations about circuits underlying prediction . Previous work identified neurons in the OT [12] and retina [6 , 8] with shifting receptive fields that account for delays in neural processing . Leonardo and Meister ( 2013 ) further showed that decoding a population of such responses with a PV can predict a moving target position . Our work shows that this type of network computation can be optimal and capture the statistics of a dynamic target . This work shows that a non-uniform population code model with a PV decoder can implement Bayesian inference for stationary and moving sources . The non-uniform population code model proposes that a prior distribution is encoded in the distribution of preferred stimuli and that the statistics of the sensory input are encoded by the pattern of neural responses across the population [23 , 24] . Here we extend this model to show that the dynamics of a population code can represent the statistics of a dynamical system . This is an important extension of the non-uniform population code model due to the dynamic nature of ethologically relevant stimuli . We make several predictions about the receptive field shifts necessary for optimal prediction . First , we predict that neurons have receptive fields that shift towards a moving source where the shift increases with the source velocity . This prediction is consistent with observations in the OT [12] and retina [6 , 8] . We also predict that the shift is sluggish when a non-uniform prior is present . This is consistent with responses of OT neurons [12] . Our analysis also leads to several predictions that have not been tested in the auditory or visual systems . In particular , we predict an asymmetry in the shifts of receptive fields for sources moving into and out of the center of gaze when a prior emphasizes the center of gaze ( Fig 6 ) . We also predict that for noisier stimuli , the magnitude of the shift will decrease and the receptive fields will become wider ( Fig 8A–8C ) . Finally , we predict that receptive fields should become narrower over time to reflect the accumulation of sensory information . Studies of neurons thought to support predictive behaviors have not yet investigated all of these response features predicted by our model . Bayesian theories of perception propose that neural systems represent statistical models of the environment , where the models may contain many parameters . The parameters of these models may be learned by an animal over multiple time scales . For the owl , information about the prior and the basic relationship between sound localization cues and source directions is primarily due to a combination of genetic changes over an evolutionary time scale and learning over the life of the animal [45] . There is evidence , however , that the owl adjusts to the noise level of the stimulus on a trial-to-trial basis [46] . We therefore predict that the noise-level parameter of the model is learned rapidly , leading to wider and more slowly shifting receptive fields in high noise environments . Future work is required to determine how the parameters of the model are learned in the owl’s auditory system . Previous studies have shown that a cascade model with a gain control component can produce the experimentally observed shifting receptive fields [6 , 12 , 8] . This model involves a negative feedback loop , causing the neural response at each time step to be influenced by its predecessors . This model is phenomenological , but it suggests that a recurrent network within the OT is sufficient to generate the receptive field shifts necessary for Bayesian prediction . However , neurons upstream from OT in ICx show direction selectivity [39 , 47] and it is therefore possible that shifting receptive fields originate in ICx . Furthermore , the asymmetric direction selectivity observed in ICx may possibly be explained by single-cell adaptation [39] rather than by a network effect . Therefore , the mechanism underlying receptive field shifts in OT remains an open question . Previous work has addressed inference in time using the Kalman filter [19 , 22 , 30 , 48] . While we determine how a population of neurons should respond to a moving stimulus but did not specify a mechanism for implementing the responses , these studies constructed networks to represent the Kalman filter estimate and variance as a function of time . One type of model produces a population code where the estimate of the target location is at the peak of a symmetric population response [22 , 30] . This is accomplished through a nonlinear encoding model involving divisive normalization . It is possible to read out the estimate using a center-of-mass decoder , but the model is limited to Gaussian distributions . Another model encodes the target estimate and variance using a linear probabilistic population code [48] . This model also relies on divisive normalization to implement the Kalman filter , but requires a nonlinear decoder to determine the estimated location from the activities . The model of Eliasmith and Anderson ( 2003 ) utilizes nonlinear responses and linear decoding . However , unlike the preferred direction vectors in the PV , the linear decoders are not in general equal to the preferred directions and are obtained using supervised learning . These models may be extended to consider the case of prediction , but the responses of neurons performing prediction in these schemes has not been investigated . Our model differs from the previous models in that the preferred direction at the peak of the population activity profile will not in general equal the PV estimate ( Fig 9 ) . This occurs because our model includes a non-uniform population , whereas previous models use a uniform population . An additional distinction between our model and previous models is that our predictions apply to nonlinear and non-Gaussian models . It has previously been shown that the PV performs poorly when decoding arm movements from motor cortical responses [49] . The work presented here does not conflict with this previous finding . We show that the PV will perform well in tracking and prediction when the receptive fields of the neurons encode the state dynamics with shifting receptive fields . This is not a general-purpose decoder , but rather must be used to read out the activity of a specialized population with shifting receptive fields such as those in the OT . Experimental evidence suggests that populations of neurons with response properties that are adapted to the natural statistics are important for perception and behavior . The work presented here shows how network properties tailored to the dynamics of moving prey allow for optimal Bayesian prediction by a population of neurons .
The Bayesian prediction at time k of the direction at a point n steps in the future θk+n , given the sequence of observations ITD1:k is computed from the posterior n steps in the future pk+n ( θ , ω|ITD1:k ) . To construct the posterior at time k+n we first compute the posterior at the current time step pk ( θ , ω|ITD1:k ) , then predict n steps in the future using the transition probability density pk+n|k ( θk+n , ωk+n|θk , ωk ) . Using the dependence relationships between direction , velocity , and ITD indicated in Fig 1 , the posterior at time k+n is given by pk+n ( θ , ω|ITD1:k ) =∬pk+n|k ( θ , ω|θk , ωk ) pk ( θk , ωk|ITD1:k ) dθkdωk . The Bayesian prediction of the direction of the sound source at time k+n conditioned on the observations ITD1:k is the mean of the predictive posterior over direction pk+n ( θ|ITD1:k ) . This posterior is found by marginalizing pk+n ( θ , ω|ITD1:k ) over the angular velocity ω . Because we are estimating a circular variable , the Bayesian prediction is the direction of the Bayesian prediction vector , defined as the vector that points in the direction of the mean value of direction n steps in the future: BVk=∫u ( θ ) pk+n ( θ|ITD1:k ) dθ , where u ( θ ) is a unit vector pointing in direction θ . We used the Kalman filter to compute the Bayesian prediction for simulations where the linear approximation to the relationship between direction and ITD was valid . The Kalman filter computes the mean and covariance of the posterior when the system is linear with Gaussian noise [41] . Given that the relationship between azimuth and ITD is nearly linear for the frontal hemisphere , a linear model is a reasonable approximation to our system . The dynamical system for the moving source can be described as: xk=Axk−1+ςk where the state vector consists of the direction and angular velocity xk=[θkωk] , the matrix A=[1Δt01] describes the state dynamics , and the noise vector contains the noise for direction and velocity ςk=[ηkνk] . The noise at time k ≥ 1 is Gaussian with zero mean and covariance matrix Q and is uncorrelated across time . The output of the system is a linear approximation to the mapping from direction to ITD plus noise: ITDk=Cxk+ξk where C = [2 . 67 0] and ξk is a Gaussian noise process with zero mean and variance R that is referred to as the observation error . The Kalman filter is used to compute the mean and covariance of the posterior at each time . Define x^i|j and Σi|j to be the mean and covariance , respectively , of the posterior at time i given observations up to time j . The mean of the posterior distribution is computed recursively through a process of prediction and updating . The prediction one step ahead in time is computed as x^k|k−1=Ax^k−1|k−1 Σk|k−1=AΣk−1|k−1AT+Q . Updating the estimate with a new observation is computed as x^k|k=x^k|k−1+Lk[ITDk−Cx^k|k−1]and Σk|k= ( I−LkC ) Σk|k−1 where the Kalman gain is Lk=Σk|k−1CT[CΣk|k−1CT+R]−1 . When an estimate has been made for the state x^k|k , it is possible to use that estimate as a basis for predicting future states at time k+n . This requires the estimate at time k to be multiplied by the state transition matrix n times: x^k+n|k=Anx^k|k . The covariance of the posterior at time k+n is computed as Σk+n|k=∑m=1nAm−1Q ( Am−1 ) T+AnΣk|k ( An ) T . We used a particle filter to compute the Bayesian prediction for simulations where the linear approximation to the relationship between direction and ITD was not valid . Particle filtering algorithms are sampling-based approaches to approximating the posterior distribution that are valid for nonlinear and non-Gaussian models [40] . The particle filter algorithm we used was adapted from [49] . The algorithm is given by the following steps: The neural population model consists of 5000 Poisson neurons with receptive fields that shift according to the prediction given in the proposition proved in the results . The preferred directions of the neurons were drawn from the prior Gaussian distribution with mean zero and standard deviation 23 . 3 deg . These preferred directions match the model of Fischer and Peña ( 2011 ) . To generate the neural responses to a sequence of ITD inputs we first computed the predictive posterior pk+n ( θ , ω|ITD1:k ) as described above . We then used our main result specifying that the activities are proportional to the ratio of the posterior and prior to generate the spiking probabilities for the population of neurons . We scaled the ratio of the posterior to the prior so that firing rates would be approximately 10 spikes/s for neurons with peak responses . Spike counts were generated for the population at each time step using independent Poisson neurons with the specified rate . The direction of the PV was used to estimate the predicted source direction at each time . The PV was tested for counterclockwise source trajectories with initial directions covering -180 deg to 180 deg in 10 deg steps and angular velocities ranging from 0 deg/s to 150 deg/s in 25 deg/s steps . We calculated the RMS error between the PV estimate θPVA ( t ) and the Bayesian prediction θBayes ( t ) to quantify the approximation error where RMS=1T∫0T ( θBayes ( t ) −θPVA ( t ) ) 2dt . | Many behaviors require predictive movements . Predictive movements are especially important in prey capture where a predator must predict the future location of moving prey . How sensory information is transformed to motor commands for predictive behaviors is an important open question . Bayesian statistical inference provides a framework to define optimal prediction and Bayesian models of the brain have received experimental support . However , it remains unclear how neural systems can perform optimal prediction in time . Here we use a theoretical approach to specify how a population of neurons should respond to a moving stimulus to allow for a Bayesian prediction to be decoded from the neural responses . This provides a novel theoretical framework that predicts properties of neural responses that are observed in auditory and visual systems of multiple species . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] | [] | 2015 | Optimal Prediction of Moving Sound Source Direction in the Owl |
Genome-wide association studies in human populations have facilitated the creation of genomic profiles which combine the effects of many associated genetic variants to predict risk of disease . The area under the receiver operator characteristic ( ROC ) curve is a well established measure for determining the efficacy of tests in correctly classifying diseased and non-diseased individuals . We use quantitative genetics theory to provide insight into the genetic interpretation of the area under the ROC curve ( AUC ) when the test classifier is a predictor of genetic risk . Even when the proportion of genetic variance explained by the test is 100% , there is a maximum value for AUC that depends on the genetic epidemiology of the disease , i . e . either the sibling recurrence risk or heritability and disease prevalence . We derive an equation relating maximum AUC to heritability and disease prevalence . The expression can be reversed to calculate the proportion of genetic variance explained given AUC , disease prevalence , and heritability . We use published estimates of disease prevalence and sibling recurrence risk for 17 complex genetic diseases to calculate the proportion of genetic variance that a test must explain to achieve AUC = 0 . 75; this varied from 0 . 10 to 0 . 74 . We provide a genetic interpretation of AUC for use with predictors of genetic risk based on genomic profiles . We provide a strategy to estimate proportion of genetic variance explained on the liability scale from estimates of AUC , disease prevalence , and heritability ( or sibling recurrence risk ) available as an online calculator .
Genome-wide association studies in human populations have facilitated the creation of genomic profiles which combine the effects of many associated genetic variants to predict risk of disease . Genetic testing has long been available for Mendelian genetic diseases for which variants within one gene are directly responsible for the disease . In contrast , the etiology of complex genetic diseases , such those listed in Table 1 , comprises both genetic and environmental risk factors . Results from genome-wide association studies have provided empirical evidence that very few associated genetic variants with effect size greater than odds ratio of 1 . 5 exist [1] , [2] . Reconciliation of these effect sizes with the , often sizeable , estimates of heritability for many complex diseases ( Table 1 ) means that we must expect there to be many ( perhaps thousands ) of genetic variants underlying complex disease if the effect size of any one variant is very small . It follows that each individual will carry a different , probably unique , portfolio of risk alleles . Whereas common risk variants have size too small to be used individually as risk predictors , profiles based on many associated genetic variants could provide useful predictions of genetic risk [3] , [4] . We define genetic risk as the risk of disease given an individual's unique multi-locus genotype; genetic risk remains unchanged throughout an individual's lifetime and so could be predicted at birth prior to exposure to many environmental risk factors . Indeed , such risk predictions could be age specific , for example , risk of type 2 diabetes at 10 years , 20 years or 50 years if genomic profile sets based on empirical data were available for these scenarios which have age-specific genetic epidemiologies . As more variants are identified in the coming years , there will be increasing interest in the prospects of genomic profiling . It has been argued that genomic profiles should be assessed in terms of their clinical validity as diagnostic classifiers [5] , [6] . The receiver operator characteristic ( ROC ) curve [7] is a well established tool for determining the efficacy of clinical diagnostic and prognostic tests in correctly classifying diseased and non-diseased individuals and has been used in the context of genomic profiling e . g . , [6] , [8] , [9] . While the area under the ROC curve ( AUC ) is an important measure for clinical validity it does not tell the whole story as it does not differentiate between the accuracy with which the genomic profile predicts the true genetic risk of individuals and the accuracy with which true genetic risk predicts disease status , which is not under our control . We believe that the ability to differentiate between these components ( i . e . the distinction between prediction of genotype and phenotype ) is important for interpretation of the value of the genomic profile , particularly as the use of genomic profiles is very much in its infancy at present . Our knowledge of the genetic epidemiology of a disease means that we can know a priori that genomic profiles might not , on their own , be accurate diagnostic classifiers . For this reason , genomic profiles should judged in the first instance on the basis of their analytic validity [10] as predictors of genetic rather than absolute risk . Of course , in the long term genomic profiles can be combined with environmental risk factors to predict absolute risk in the context of clinical utility . Genomic profiles should improve upon family history which has long been used as a crude estimate of genetic risk ( see Text S1 ) . In this paper , we provide insight into the genetic interpretation of AUC . We begin by considering quantitative traits for which the concepts of accuracy of risk prediction are well developed . For disease traits we differentiate between measures on the observed scale of disease versus the underlying scale of disease risk as we believe recognition of scale of measurement is often overlooked . We define AUCmax as the maximum AUC that could be achieved for a disease when the test classifier is a perfect predictor of genetic risk . We quantify the relationship between AUCmax and heritability of liability and disease prevalence ( lifetime morbidity risk ) . We show how to interpret AUC ( which is a measure on the observed disease scale ) of a genomic profile as the proportion of variance explained ( or accuracy of prediction squared ) on the underlying liability scale . Finally , we benchmark the value of genomic profiles by comparing them to the AUC expected when family history resulting from shared genetic risk factors is used as a predictor of genetic risk .
For quantitative traits , in which phenotypic scores are ( or can be transformed to be ) normally distributed , the efficacy of a genomic profile is naturally expressed as the proportion of the genetic variance explained by the profile . The variance in phenotypes , VP , can be partitioned into variance of genetic values , VG , so that the proportion of the variance that is genetic is the heritability VG/VP . Genomic profiling provides a direct estimate , , of true genetic values , G , for individuals in a population and the efficacy of a genomic profile can be expressed as the proportion of the genetic variance explained by the profile /VG . We define = /VG , since in selection theory [11] , used in livestock and plant breeding , the correlation between predicted and true genetic risk ( ) is used as the measure of accuracy of prediction , , and if the predictor is unbiased ( the regression of G on is 1 ) , . The ratio /VP is estimated as the R2 from the regression of P on and is interpreted recognising its upper limit to be VG/VP or heritability . These measures show that for quantitative traits , the accuracy with which the genomic profile predicts genetic risk is clearly separable from accuracy with which the true genetic risk predicts the phenotype . In contrast , AUC is a measure of the efficacy with which predicts phenotype which , as shown below , has an upper limit constrained by the heritability , and also prevalence , of the disease . For disease traits , the phenotype has two possible values , either affected or not affected . On this observed scale , the directly measurable genetic parameters are those of recurrence risks to relatives , λR for relatives of type R , which is the ratio of the prevalence of disease in the relatives of affected individuals ( KR ) compared to the prevalence in the population ( K ) , where cov ( X , R ) the covariance in disease status between diseased individuals X and their relatives on the observed disease risk scale [12] . For example , when the relatives are monozygous twins ( R = MZ ) , Cov ( X , MZ ) = the genetic variance , with the subscript “01” denoting the all-or-none disease risk scale . On this scale , the majority of the genetic variance is non-additive , especially when disease prevalence is low [13] , [14] . The broad sense heritability on this scale is = ( λMZ -1 ) K/ ( 1-K ) where λMZ is the monozygotic twin recurrence risk , assuming there is no common environmental component to the recurrence risk . is not a normally reported statistic because of its dependence on disease prevalence [15] . If the relatives are siblings ( R = S ) then λS is the sibling risk ratio and Cov ( X , S ) = [11] , where the variance subscripts A and D denote additive and dominance terms , and in combination denote epistatic variance terms . Thus , although λS is an estimable quantity , it is not simply related to the genetic variances on the observed binary scale . The genetic properties of disease are much more easily understood by using the threshold liability model [11] , in which risk of disease is transformed to a normally distributed liability scale P ∼N ( 0 , 1 ) and P = A + E , where A∼N ( 0 , ) are the genetic effects on the liability scale . On this scale the genetic effects combine in an additive way; is the narrow sense heritability on the liability scale ( or heritability of liability ) and on this scale broad sense and narrow sense heritability are equal . E are independent environmental effects , E∼N ( 0 , 1- ) . The biological plausibility of an underlying normally distributed liability to disease is based on the assumption that complex traits are influenced by many variables; the central limit theorem states that the distribution of the sum of independent random variables approaches normality as the number of variables increases . Under the threshold liability model individuals are affected when P >T , where T is the threshold on the normal distribution which truncates the proportion of affected individuals or disease prevalence ( i . e . , K ) , T = Φ−1 ( 1-K ) , Φ ( T ) = 1-K , where Φ ( T ) is the cumulative density function of the normal distribution up to values of T , e . g . if K = 0 . 05 , T = 1 . 645 . The threshold liability of risk scale has much nicer properties than the observed disease scale and provides a framework for comparison of scenarios independent of disease prevalence . The relationship between heritability of liability and the directly estimable parameters of K and λS is ( 1 ) [16] with and z the height of the standard normal curve and T1 = Φ−1 ( 1- λS K ) , i . e . the threshold T1<T when λS>1 , reflecting that the prevalence amongst sibs of affected individuals , KS is greater than the prevalence in the population as a whole ( e . g . if K = 0 . 05 and λS = 2 , z = 0 . 103 , T1 = 1 . 282 , = 0 . 371 ) . The AUC is a statistic calculated on the observed disease scale and is a measure of the efficacy of prediction of phenotype using a test classifier . The ROC plots the true positive rate ( TPR or sensitivity ) against the false-positive rate ( FPR or 1-specificity ) . TPR = probability ( positive test result|diseased ) and FPR = probability ( positive test result|not diseased ) . Since these probabilities are conditional , they are not dependent on the number of cases or controls tested , except through the sampling variance associated with them . In genomic profiling the ROC is obtained by ranking a set of individuals with known disease status by their genomic profile from lowest estimated risk ( i . e . , profile score ) to highest estimated risk and then assessing sensitivity and specificity assuming a cut-off after each rank ( starting with the highest ranked individual ) . If nd and nd' are the numbers of diseased and not diseased individuals , and if the individual with the highest predicted genetic risk has rank r1 = nd + nd' = n , AUC can be calculated directly from the mean rank of the diseased individuals ( ) , ( 2 ) ( see example in Figure S1 ) . Equally , AUC can be calculated as AUC = 0 . 5 ( 1 + D ) where D is the Somers' rank correlation [17] between risk profile and disease status ( 1 = diseased , 0 = not diseased ) . Another equivalent definition of AUC is the probability that a randomly selected pair of diseased ( d ) and non-diseased ( d' ) individuals are accurately classified [18] . The probability is the same as the probability that difference between the genetic liability of the d and d' individuals is greater than zero . This difference is approximately normally distributed with mean μd - μd' and variance . Using the liability threshold model and results of standard genetic selection theory [11] the means ( μ ) and variances ( σ2 ) of the genetic liability of d and d' individuals arewhere v = -iK/ ( 1 – K ) . The genetic liabilities of the d and d' groups are each approximately normally distributed , the approximation being less accurate for high heritabilities . Therefore , ( 3 ) A useful property of AUC ( as discussed above ) is that for a given disease the estimated AUC is independent of the relative proportions of cases and controls in the sample being classified [7] , i . e . the mean rank is approximately the same if the proportion of cases: controls is K: ( 1-K ) or 1∶1 . Or equally , the probability of a randomly selected case and control being correctly ranked is independent ( except for sampling ) of the number of cases and controls measured . We can use equation 3 to estimate the variance on the liability scale explained by a genomic profile , x , by making the subject of the equation , but renaming it as , recognising that it represents the proportion of variance explained by the profile . Then , from two measurable parameters , K and AUC , we can calculate , ( 4 ) Where Q = Φ−1 ( AUC ) . From this , we can calculate the proportion of the known genetic variance explained by the genomic profile ( 5 ) using the estimates of K and λS to calculate ( equation 1 ) . We can also calculate the proportion of the sibling risk explained by the profile , ( λS[x] – 1 ) / ( λS – 1 ) , where λS[x] = ( 1-Φ ( T1[x] ) ) /K and ( 6 ) [19] . and ( λS[x] – 1 ) / ( λS – 1 ) measure the same concept but in different ways and on different scales; both are useful criteria for assessing the extent to which the genomic profile accounts for the known genetic component of disease . We consider family history as a predictor of genetic risk in the Text S1 . We used simulation under the liability threshold model [11] , [14] to check our derivations . We simulated 100 , 000 nuclear families sampling risk on the liability scale , P = A + E , A ∼ N ( 0 , ) for parents , and A = ½Adad+½Amum+Amend for children , where the Mendelian segregation terms were random numbers sampled as Amend ∼ N ( 0 , ½ ) ; E ∼ N ( 0 , 1 - ) . Individuals were considered affected , P01 = 1 , if P >Φ−1 ( 1-K ) = T , otherwise individuals were not affected and P01 = 0 . Genetic values on the observed scale , G01 , were calculated as the normal probability , . From this we could calculate , , ( using the G01 and P01 of the first child from each family ) and sibling recurrence risk . AUCmax was calculated from the mean rank of diseased individuals using equation 2 when ranked on A .
In Figure 1A we consider two diseases both with heritability of liability , = 0 . 2 , plotting probability of disease ( i . e . G01 ) vs genetic liability ( i . e . A ) . To allow an extreme comparison , one of the diseases has prevalence K = 0 . 5 and the other , K = 0 . 01 . Figure 1B also considers two diseases with prevalences K = 0 . 5 and 0 . 01 , but in this case both have = 0 . 8 . In Figure 1A and 1B , the position of the rise in probability of disease along the x-axis reflects the disease prevalence and the steepness of the rise reflects the heritability of the disease . In Figure 1A the distribution of genetic liabilities on the underlying scale is exactly the same for these two diseases , but when K = 0 . 01 higher genetic liabilities are needed before probability of disease rises above virtual zero ( virtual because it is not exactly zero , but very close to zero ) ; similarly for the diseases in Figure 1B . Figure 1C and 1D plot the ROC curves for the diseases considered in Figure 1A and 1B , respectively . These graphs demonstrate firstly ( not unexpectedly ) , that for diseases with the same prevalence , genetic liability is a better predictor of disease status for diseases with higher heritability and secondly , that for diseases with the same heritability , genetic liability is a better predictor of disease status for rarer diseases , because a higher proportion of those with high genetic liability are actually diseased . For example , if we used genetic liability of ≥1 as our predictor of disease , then the TPR is 0 . 26 and the FPR = 0 . 00 , when K = 0 . 5 , compared to TPR = 0 . 99 and the FPR = 0 . 12 , when K = 0 . 01 . These graphs demonstrate that maximum value of AUC ( i . e . AUCmax ) when the test classifier is a genetic predictor is dependent on both and K . Figure 2 plots AUCmax vs , for K = 0 . 001 , 0 . 01 , 0 . 1 , 0 . 3 from simulation ( dashed line ) and from equation 3 ( solid line ) and shows that AUCmax is particularly constrained for more common or low heritability diseases . Jannsens et al [3] , in their Fig . 4 , have shown the relationship between AUC and the proportion of variance on the disease scale explained by the genomic profile; since their genomic profile assumed all genetic variants were known without error their graph represents the relationship between AUCmax and . Our simulation results provided the same relationship when plotted on this scale ( Figure 3 , solid line ) . In Figure 3 we show the relationship of AUCmax with and ( for each simulation combination of K and , the AUCmax and are calculated ) . Table 1 lists AUCmax for a range of complex genetic diseases calculated using equation 3 , with calculated using equation 1 from published estimates of K and λS . Despite being observable , the parameters K and λS are subject to considerable sampling variance; we have tried , where possible , to take estimates from reviews or large studies , but large study samples simply do not exist for some low prevalence disorders . The values of AUCmax show that it should be possible for a genomic profile for complex diseases to exceed 0 . 75 , the threshold regarded [20] as making a diagnostic classifier clinically useful when applied to a sample considered to be at increased risk . However , based on the results in Table 1 only the diseases with high heritability and low prevalence , such as Type I diabetes , Crohn's Disease and Lupus , can achieve an AUC , by genomic profiling alone , above the 0 . 99 threshold regarded [20] as being required for a diagnostic classifier to be applied in the general population . In Table 1 , we also consider the AUC expected under scenarios where a genomic profile accounts for only a half ( AUChalf ) or a quarter ( AUCquar ) of the known genetic variance . These results show that for rare diseases genomic profiles can be useful classifiers of disease ( AUC>0 . 8 when K<0 . 01 ) , when the profile explains only a quarter of the genetic variance . Using equations ( 4 ) and ( 5 ) we calculate for the diseases listed in Table 1 when AUC = 0 . 75 . The results ( Table 1 ) show that the same AUC can represent quite different successes of the genomic profile in representing the known genetic variance , ranging from 0 . 10 to 0 . 74 . If we are able to explain half of the known genetic variance with identified risk variants then genomic profiles for most complex genetic disease ( AUChalf , Table 1 ) will achieve some clinical validity as AUC is >0 . 75 for all but bladder cancer , for the examples provided . Consider the first listed example in Table 1 , age related macular degeneration ( AMD ) . Based on the review of Scholl et al [21] and the large twin study of Seddon et al [22] we have used a prevalence after 80 years age of advanced AMD K = 11 . 8% and a sibling recurrence risk representing the genetic contribution of λS = 2 . 2 , which correspond to heritability on the liability scale of = 0 . 68 ( equation 1 ) . If the genetic test explains all the genetic variance ( = 1 ) , the maximum AUC that could be achieved by a genomic profile is AUCmax = 0 . 92 . If only half or a quarter of the genetic variance can be detected by genomic markers then the maximum AUC that can achieved are AUChalf = 0 . 81 and AUCquar = 0 . 72 , respectively , values that exceed the prediction of genetic risk based of the most optimistic scenario from a prediction based on family history ( Text S1 ) . If complete disease status is known for all siblings , parents , grandparents , aunts , uncles and cousins then the maximum AUC that could be achieved is 0 . 71 , translating to a genomic profile that explains 0 . 21 of the genetic variance ( Table S1 ) . In practice , the AUC for a risk predictor based on rs1061170 a single nucleotide polymorphism in the complement factor H ( CFH ) gene was 0 . 69 [23] ( and was approximately equal for advanced AMD cases vs controls and all AMD cases vs controls ) . From equations 4–6 , = 0 . 12 , λS[x] = 1 . 17 , = 0 . 17 and ( λS[x] – 1 ) / ( λS – 1 ) = 0 . 15 .
The AUC is a widely used statistic that summarises the clinical validity of a diagnostic or prognostic test . However , the AUC statistic of a genomic profile alone has an upper limit ( i . e . AUCmax ) which depends on the genetic epidemiology of the disease , namely the disease prevalence and heritability . It is important that in the first instance , particularly when genomic profiling is in its infancy , that genomic profiles are judged on their ability to predict genetic risk ( their analytic validity ) rather than on the basis of clinical validity [10] . Since AUC is estimated as a function of a rank correlation its genetic interpretation is not immediately obvious . Here we provide a genetic interpretation of the AUC expressed in terms of it genetic epidemiology parameters ( equation 3 ) . A relationship between AUCmax and heritability was first demonstrated graphically by Janssens et al [3] ( see solid line Figure 3 ) . However , their representation was of broad sense heritability on the observed scale ( i . e . ) which is a little used measure of heritability because of its dependence on disease prevalence [13] . Here we show ( Figure 2 and equation 3 ) the relationship between AUCmax and the more commonly used measure of heritability , the heritability of liability ( i . e . , ) We show that AUCmax is dependent on both and disease prevalence ( i . e . K ) . Initially , it may seem counter-intuitive that AUC depends on disease prevalence since for an individual disease TPR and FPR are independent of the proportion of cases and controls measured and therefore of the sample prevalence . However , as we have clearly shown ( Figure 1A and 1B ) the dependence on disease prevalence results from our ability to generalise across diseases in the context of a test classifier being a genomic profile . In contrast to our results and those of Janssens et al [3] , Clayton [24] provided an expression for ROC under a polygenic model which is independent of population disease prevalence . His derivation assumes that the effect of each locus is additive on the log risk scale [25] . Slatkin [26] and we [27] have found that this model allows probabilities of disease that exceed one , which although they occur with low frequency can have substantial impact on the estimates of recurrence risk and genetic variance . Under this model there is a relationship between recurrence risk to monozygotic twins and to siblings of λMZ/ = 1; this ratio is not achieved when probabilities of disease are constrained to their natural parameter space of a maximum of 1 . Furthermore , empirical estimates of the ratio of λMZ/ from the studies listed in Table 1 that provide estimates of λMZ and λS are mostly less than 1 . 0 [27] , particularly for low prevalence diseases . Recognising that these estimates are subject to sampling variance , the estimates of λMZ/ are 1 . 1 ( AMD ) , 0 . 4 ( coronary artery disease ) , 0 . 8 ( breast cancer ) , 0 . 7 ( schizophrenia [25] ) , 0 . 9 ( rheumatoid arthritis ) and 0 . 4 ( Type I diabetes ) . Therefore , we believe the model used by Clayton to derive the relationship between AUC and heritability ( or sibling recurrence risk ) independent of disease prevalence is not valid . AUC is a useful measure because of its independence of the numbers of diseased and diseased individuals tested , but we advocate the reporting of an estimate of the proportion of the known genetic variance on the liability scale ( ) or the proportion of sibling risk accounted for by the profile and we provide a method to do this using the estimated AUC , disease prevalence and heritability on the liability scale or sibling recurrence risk ( equation 5 ) . An AUC of 0 . 75 can imply anything from 0 . 10 to 0 . 74 of the genetic variance explained by the genomic profile for the complex diseases listed in Table 1 . The correlation has long been the benchmark in non-human genetics of accuracy of genetic risk predictors . can be calculated from three measurable statistics , disease prevalence , sibling recurrence risk and AUC of the profile ( using equations ( 1 ) and ( 4 ) ) . In this way , estimates of AUC can provide direct estimates of the proportion of ‘missing heritability’ [28] which takes into account the interdependence of identified associated variants . Currently , the derivation of genomic profiles is very much in its infancy . As the sample size of genome-wide association studies increase , we can expect genomic profiles to include more and more validated associated variants . However , is constrained by the variance that could be detected by the markers that are genotyped recognising that the current generation of genome-wide chips explain at most ∼80% of the known variance in single nucleotide polymorphisms across the Caucasian genome [29] . This , in turn , may only be a fraction of the total genomic variance once structural variants such as copy number variants are included [30] . The actual variance explained by the profile depends on the sample size ( i . e . , power ) of the studies from which associated genetic variants have been detected . It is likely that there are many variants which have such a small effect size that they will be impossible to detect even with very large samples . Although each such variant makes only a very small contribution to the genetic variance , there may be so many that a sizeable proportion of the variance will go undetected . Even if only quarter of the genetic variance is detectable by our future genotyping technology , the AUC is still greater for the genomic profile than for family history ( ignoring shared environmental risks of family members , Text S1 ) . In our derivations we have assumed the liability threshold model [11] , [14] . Slatkin [26] demonstrated that the threshold model was one of several genetic models that provided the necessary steep increase in probability of disease with increasing load of genetic risk alleles [26] . The main assumption of the liability threshold model is that the distribution of liability scores is unimodal which should be achieved as long as there is no single unidentified genetic or environmental of very large effect [11] . The model accommodates any distribution of risk allele effect sizes and risk allele frequencies as long as there are sufficient ( “more than one or a few” [11] ) risk alleles in the population to create an approximately normal distribution of genetic liability scores . Since our simulation results of AUCmax vs ( Figure 3 ) based on the liability threshold model agree with those of Janssens et al [3] who used a logit model to combine genetic risks from individual genetic variants , it is clear that the dependence of AUCmax on heritability and disease prevalence is not a function of the threshold model . We have also assumed that a genetic profile is applied in the same “average” environment as the genetic risks were estimated and we have assumed that all familiality is of genetic origin . The AUCmax will be lower than those derived here if any part of the sibling recurrence risk reflects co-variation of non-genetic origin . Using recurrence risks from different types of relatives , the importance of common environmental factors can be assessed and a λS which reflects the genetic contribution of sibling recurrence can be used in our calculations . We have also assumed that the genomic profile consists of genetic markers associated with disease that are passed on according to the rules of Mendelian inheritance . In the future , a genomic profile might include non-heritable genetic variants , for example recurrent de novo copy number variants or perhaps methylation status variants ( for which the inheritance pattern , if any , is currently unclear [31] ) . Such variants , although genetic , do not contribute to the similarity between relatives , and so would be included in the environmental component when partitioning variance . Under these circumstances it is possible that a genomic profile could exceed the AUCmax based on sibling recurrence ratio . Our calculations assume that we know the population parameters K and λS ( and therefore ) . Estimates of these parameters are sometimes based on small sample size and are subject to sampling bias or different definitions of the disease . In particular , prevalence rates can depend on the age distribution of the population in which they are measured . In addition , recurrence risk ratios of relatives have a maximum possible value which is dependent on the disease prevalence , so that higher risk ratios are achievable when disease prevalence is lower; and estimates of sibling risk ratio and disease prevalence calculated in different studies sometimes reflect this dependence . In Table 1 , we included two different estimates for both schizophrenia and bipolar disorder , but for these examples the estimates of AUCmax are robust to the magnitude of differences reported in genetic epidemiology parameters for individual diseases . At present , genomic profiles based on validated associated variants do not come anywhere close to the maximum implied by their AUCmax; Jakobsdottir et al [6] have reported AUC of 0 . 80 for risk of cardiovascular events , 0 . 64 for type 2 diabetes , 0 . 56 for prostate cancer , 0 . 66 for Crohn's Disease and 0 . 79 for age related macular degeneration . This is not surprising given the effect size of individual associated variants discovered in genome-wide association studies , which imply that much larger sample sizes will be needed to discover the majority of the variants that explain the genetic variance [4] . However , already these genomic profiles outperform family history ( resulting from shared genetic risk only ) for four out of five of these diseases . Although the AUC is a useful summary statistic for clinical validity , in practice clinical utility depends on many other factors such as the benefits versus risks of the intervention strategies that follow from the risk prediction [5] , [32]; these important factors are not considered here . We have provided a genetic interpretation of and insight into the AUC statistic calculated under a genomic profile . Time will tell if genetic variants amenable to genotyping are able to reconstruct the known genetic variance in its totality . Even if it is possible to explain only a quarter of the known genetic variance , the genomic profile will be a more useful predictor of genetic risk than self-reported family history ( in the absence of shared environmental risk factors ) which is a commonly used measure for targeted screening programmes for complex genetic diseases . In practice , predictions of risk to disease will incorporate both genetic and environmental risk factors to produce the best predictions of absolute risk to disease . Here we provide a benchmark for the expected contribution from the genetic component of the prediction illustrating that the same AUC estimated for different diseases can imply quite different proportions of genetic variance explained by the genomic profile , which is often overlooked ( e . g . [5] ) . Ultimately , genomic profiles may be used without contributions from environmental risk factors , since the contribution from the genomic profile can be estimated perinatally , prior to exposure by many environmental risk factors and when limited family history of disease is available . Indeed , one purpose of a genetic risk predictor is to allow individuals to choose to modify their exposure to environmental risks . We provide a simple online calculator ( http://gump . qimr . edu . au/genroc ) to calculate i ) the maximum AUC for a genomic profile of a disease given estimates of disease prevalence and sibling recurrence risk or heritability of liability , ii ) the proportion of variance explained on the liability scale given an estimate of AUC from a risk predictor and disease prevalence and iii ) proportion of genetic variance or of sibling risk explained given an estimate AUC , disease prevalence and sibling recurrence risk [2] . | Genome-wide association studies in human populations have facilitated the creation of genomic profiles that combine the effects of many associated genetic variants to predict risk of disease . However , genomic profiles are inherently constrained in their ability to classify diseased from non-diseased individuals dictated by the genetic epidemiology of the disease . In this paper , we use a genetic interpretation to provide insight into the constraints on genomic profiles for risk prediction . We provide a strategy to estimate proportion of genetic variance explained on the liability scale from estimates of AUC , disease prevalence , and heritability available as an online calculator . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [
"genetics",
"and",
"genomics/disease",
"models",
"genetics",
"and",
"genomics/complex",
"traits"
] | 2010 | The Genetic Interpretation of Area under the ROC Curve in Genomic Profiling |
Characterizing infectivity as a function of pathogen dose is integral to microbial risk assessment . Dose-response experiments usually administer doses to subjects at one time . Phenomenological models of the resulting data , such as the exponential and the Beta-Poisson models , ignore dose timing and assume independent risks from each pathogen . Real world exposure to pathogens , however , is a sequence of discrete events where concurrent or prior pathogen arrival affects the capacity of immune effectors to engage and kill newly arriving pathogens . We model immune effector and pathogen interactions during the period before infection becomes established in order to capture the dynamics generating dose timing effects . Model analysis reveals an inverse relationship between the time over which exposures accumulate and the risk of infection . Data from one time dose experiments will thus overestimate per pathogen infection risks of real world exposures . For instance , fitting our model to one time dosing data reveals a risk of 0 . 66 from 313 Cryptosporidium parvum pathogens . When the temporal exposure window is increased 100-fold using the same parameters fitted by our model to the one time dose data , the risk of infection is reduced to 0 . 09 . Confirmation of this risk prediction requires data from experiments administering doses with different timings . Our model demonstrates that dose timing could markedly alter the risks generated by airborne versus fomite transmitted pathogens .
Microbial risk assessment models are valuable tools for estimating the risks associated with exposures to pathogens in the environment pathogens [1] . Central to this estimate is a dose-response model that predicts the probability of infection given a dose exposure magnitude . In current microbial risk assessment models dose accumulates over time and the probability of infection is based on the total accumulated dose over that period of time [2]–[4] . This assumes that each pathogen particle carries a risk of infection that is independent of when other pathogens have arrived to a host; i . e . , three exposures to dose X generate the same total risk as one exposure to a 3× dose . We put forth an alternative dose response model that assumes the current capacity of immune effectors to control an arriving pathogen should be affected by 1 ) how many effectors are occupied fighting previously or simultaneously arriving pathogens , 2 ) how many effectors have been depleted in fighting previously arriving pathogens , and 3 ) how many effector reinforcements have arrived due to usual effector turnover rates or due to a stimulus from prior pathogen exposure . If dose-timing effects arise from such immune effector dynamics , then infection-risk calculations that do not take these dose-timing effects into account could lead to errors . For example , errors could arise in models of influenza transmission as follows . Pathogens arriving to a host via aerosols do so more frequently but at lower doses than pathogens arriving via hand or fomite mediated inoculations . Models of influenza transmission that do not account for dose-timing effects , such as the model by Atkinson and Wien [4] , might misdirect influenza control resources to masks from hand hygiene . Models that assume independent single dose effects will require more extreme cleaning to reduce risks to acceptable levels than models capturing immune effects on dose timing . Evaluating the potential importance of such dose-timing effects is difficult for two reasons . First , immune control of pathogens is complex; not enough detailed knowledge regarding that complexity is available to provide a high degree of confidence in a-priori causal model predictions . Second , there is almost no direct observational data documenting the presence or absence of dose-timing effects . Although various studies have given pathogen exposure doses over time [5]–[10] , only Brachman et al . [11] , has been conducted in a manner that allows one to calculate risks for comparable doses administered over different temporal windows . In this paper we have taken an approach intended to stimulate science that will address both of these issues . We develop a simple model that illustrates the need to generate new data that can describe dose-timing effects while at the same time providing a base upon which to build more realistic models that incorporate more data and theory on immunity . Our model addresses immune control of pathogens between the time pathogens arrive at a host and the time they are either eliminated or have multiplied enough so that an acquired immune response will be needed for control . We make our model general enough to capture dynamics of pathogen control that might arise from established antibodies and T-cells , macrophages , polymorphonuclear leukocytes , plasma cells , dendritic cells , complement cascades , chemokines , interleukins , interferons , toll like receptors , and other diverse elements affecting immunity . But we lump all these mediators of pathogen control into a highly abstract entity we label as immune effectors . We assume that the dynamic effects of limited immune effector numbers are similar whether the limitation arises from immune effectors being occupied with previously arrived pathogens or from prior consumption of immune effectors in their process of killing pathogens . Therefore we only model the latter source of immune effector limitations . The resulting model is one where any single pathogen always has some chance of initiating an infection but the risk of infection associated with each additional pathogen exposure can markedly increase at higher pathogen doses given over short temporal windows . The exact dynamics of our model will vary as realistic details are added to it . Our goal here is simply to illustrate the importance and inevitability of immune mediated dose-timing effects so as to stimulate further empirical and theoretical work . The structure of the paper is as follows: in the methods section we describe the Cumulative Dose model and analyze its dynamics . In the results section we use the Cumulative Dose model to fit experimental data assuming a fixed temporal exposure window to simulate the archetypical single dose experiment of dose-response trials . Using the estimated model we show the effect of changing the length of the temporal exposure window . Finally , the conclusions and future research are presented in the discussion section .
The model is based on a stochastic population of individual pathogens and immune effectors . Since the focus of our analysis is how small populations of pathogens either die out or lead to infection initiation , we cannot rely on the mean-field solution provided by the deterministic framework [12]–[14] . The state of the system is defined by the pair ( ) representing the number of immune effectors and the number of pathogens , in any single host , respectively . The system is defined by the following set of state transitions: ( 1 ) ( 2 ) ( 3 ) ( 4 ) The number of immune effectors can increase at: 1 ) a rate , which models the constant arrival of immune effectors regardless of the current state of the immunological system; and 2 ) a rate , which models the recruitment of immune effectors in the presence of pathogens . This term is intended to reflect cytokine induced recruitment of remote immune effectors to a pathogen invasion site and not acquired immunity . We assume that the relative endpoints of infection takeoff or pathogen elimination are reached before an acquired immune effect comes into play . Immune effectors decrease either at a natural death rate , or at a mass-action deactivation rate due to the encounter with pathogens . The number of pathogens can increase by reproduction at a rate or by arrival during the inoculation period at a rate . Here represents the net reproduction rate that aggregates birth and death rates . Pathogen numbers decrease due to interaction with immune effectors as a mass-action deactivation process at the rate . The initial state of the system is set to . No chronic low-level exposures or remaining pathogens from prior exposures are considered . The system starts from the clean state: no pathogens and the stationary number of immune effectors in the absence of pathogens . The inoculation process is characterized by the dose of exposure and the temporal exposure length ; i . e . , the dose that is composed by pathogens is inoculated into the host during a period of time units . Therefore , the arrival of external pathogens is modeled as the rate during the inoculation period . Once inoculation has finished the pathogen arrival rate becomes zero . Thus , the rate depends on time and is defined as During , the pathogens , arrive over a continuous time in the presence of the immunological response to those pathogens . Once the inoculation has finished , only the immunological response remains . We set the unit of time to an hour . That keeps us in the range where we think exposure fluctuations are making a difference and out of the range where adaptive immune system feedbacks come into play . Due to stochastic effects and the fate of a relatively small population of pathogens and immune effectors , the same inoculation dose administered in the same time frame does not necessarily have the same outcome . Each replication ( i . e . run ) of the model corresponds to a dose trial on a new subject . All the numerical results are the average of 104 runs of the Cumulative Dose model implemented with the Gillespie algorithm [15] using C . The criteria to stop the simulation is either extinction of pathogens after the inoculation period ( ) or pathogens diverging to a very large number , , corresponding to no infection and infection respectively . The probability of infection for a pair is the proportion of simulations that diverge to a large number as opposed to equilibrating to the state of no pathogens . Figure 1 illustrates the stochastic process effects on pathogen dynamics given a fixed time of exposure for different inoculation doses . The main plot in this figure is the time course of the number of pathogens for 100 independent dose trials given a dose of 60 pathogens administered over one unit of time . The number of pathogens steadily grow during the inoculation period , from 0 to 1 , since the rate of arrival of pathogens ( ) is much faster than immunological killing of pathogens . Once the entire dose has been inoculated at time = 1 , the external arrival of pathogens stop ( ) and the immunological response dominates the rest of the dynamics . In this particular case , the population of pathogens becomes extinct in 33 cases out of 100 , thus , the probability of infection given a dose of 60 pathogens over 1 unit of time is 0 . 67 . Analogously , for a dose of 25 the probability of infection is 0 . 02 and for a dose of 90 the probability of infection is 0 . 98 ( insets of Figure 1 ) . Figure 1 illustrates how the Cumulative Dose model yields higher probability of infection when the inoculated dose is increased . The length of time over which the dose is administered , , also plays a crucial role in the probability of infection . At one extreme where all the pathogens were inoculated at once ( ) , the immune system has no time to react , and the initial state of the system is . From this initial state , the immunological response dynamics determines the fate of the pathogens: either extinction or unbounded growth of pathogens diverging towards infinity . For , however , the initial state after all pathogens have been inoculated ( ) is not the expected , but rather a distribution of probabilities over the space of possible states . Figure 2 shows the stochastically determined distribution of system states at the point in Figure 1 where the exposure time has just ended . It illustrates the effect of different temporal exposure lengths , ranging from ( six minutes ) to Te = 50 hours . Panel B shows this point of time for the settings in Figure 1 . The longer the exposure length , the larger will be the variance in the distribution of probabilities . Furthermore , a longer exposure length also affects the average state after inoculation . Both the pathogen levels and the immune effector levels decrease from the instantaneous inoculation values as the exposure window length increases . But the balance between these increasingly favors the immune effectors . Longer temporal exposure lengths dilute the arrival rate of external pathogens , . Consequently the immunological response has more time to neutralize the existing pathogens before the arrival of new pathogens . On the other hand , as the temporal exposure lengths decrease , an increased number of immune effectors are consumed in killing pathogens , leading to a higher probability of unbounded growth of pathogens , and thus infection . For and the average state after inoculation is very close to the ideal instantaneous inoculation , . To better understand the dynamics once inoculation is over , we included the numerically calculated separatrix as if the system were deterministic ( red-dashed line in Figure 2 ) . Although this separatrix is only truly valid for the analogous deterministic model , it indicates the probable fate of different initial states . For the deterministic system , the separatrix separates those states that go to infection from those that do not ( see subsection on Deterministic Analysis ) . As temporal exposure length increases , the distribution of probabilities gravitates towards the space of states that go to no-infection ( below the separatrix ) . Further understanding of the stochastic dynamics of the Cumulative Dose model can come from a deterministic description of the system that assumes a continuous large number of immune effectors and pathogens . We focus our analysis on the dynamics after the inoculation period , so is set to 0 and removed from the equations . This analysis on the deterministic version helps illustrate the interactions between pathogens and immune effectors that result either in infection or extinction of pathogens . The stochastic system is fully described by a multivariate master equation [16] , which can be expanded in a deterministic formulation known as macroscopic law . The deterministic version of the cumulative dose model is as follows , ( 5 ) ( 6 ) where and are continuous variables of the population of pathogens and immune effectors respectively . The fixed points of the deterministic version of the cumulative dose model are where the pathogen has been eliminated and immune effectors are in equilibrium and where the forces of pathogen growth are balanced by immune dynamics affecting pathogen death . Note that in the stochastic analyses of this model as in Figure 1 , this point is never reached . Instead simulations are terminated when growth takes off toward this point . A simple analysis of the stability of the fixed points reveals the space of parameters in which the solution is well-defined . The point is the equilibrium of no infection—the equilibrium of the system in the absence of pathogens . When the system gravitates towards the immunological system prevents pathogens from growing , resulting in pathogen extinction and therefore no infection . To evaluate the stability of the fixed point , we formulate the Jacobian matrix of the system of equations on . ( 7 ) For a stable equilibrium , both Eigenvalues of the Jacobian matrix need to be negative , or equivalently , the matrix must have a negative trace and a positive determinant . For the trace of the Jacobian to be negative the condition must be true . Since the positive determinant condition , , is more restrictive it subsumes the condition for a negative trace . The second fixed point is only well-defined when both and are positive , since negative number of pathogens and immune effectors are impossible . The number of pathogens is only positive when . Given the condition of a positive determinant , , the sign can only be negative , consequently . Therefore , the system is well defined — i . e . has a stable equilibrium at no infection and with both fixed points in the positive quadrant — only when the following condition 8 is met ( 8 ) Once we determine the stability of we need to characterize the second fixed point . After some basic algebra , the determinant of the Jacobian matrix for can be expressed as follows: . Given condition 8 , both terms are positive , which makes the determinant negative . As a result the Eigenvalues of the Jacobian are real with different signs . Therefore , is a saddle point as shown in Figure 3 . The vector field in Figure 3 illustrates the dynamics of the cumulative dose after the inoculation period . The probability of being in a given state after inoculation is shown in Figure 2 . If the system were deterministic then we could anticipate the probability of infection by summing the probability of those states below the separatrix . This does not hold for the stochastic Cumulative Dose model . Nonetheless , the deterministic vector field , shown in Figure 3 , serves as an approximate description of what happens in the stochastic model . For instance , let us take the probability distribution of states when centered at , i . e . , and . The typical dynamic results in the decrease in number of pathogens and immune effectors , gravitating towards the saddle point , from which it will bifurcate to the stable point of no-infection , or an unbounded growth of pathogens . In the case of and , most of the states are already very low in pathogens , and consequently the number of immune effectors will eradicate the few pathogens still existing and go to the stable equilibrium of no infection . However , there is a non-zero probability , albeit small , of being in a state with a large number of pathogens and a small number of immune effectors . In this case , stochastic perturbations aside , the pathogens will keep multiplying producing infection in the host .
In this section , we fit empirical data on multiple pathogens for the single event inoculation scenario . Next , we extend our analysis to incorporate different temporal exposure windows and patterns of inoculation . The first empirical dataset to which we apply the Cumulative Dose model is Poliovirus type 1 [17] . The cohort for this experiment was 32 2-month-old infants . Inoculation was oral . Figure 4 and Table 1 show the fit alongside a fit to the Exponential model ( ) according to [18] . The cohort for the Cryptosporidium parvum study [18] was 35 healthy subjects ( 12 men and 17 women , age range between 20 and 45 years ) . The strain was an isolate from a calf and the inoculums were orally administered via capsules . Figure 5 and Table 2 show the fit alongside a fit to the Exponential model ( ) according to [20] . Finally , we tested the Cumulative Dose model against a dataset for Rotavirus [19] . The cohort for rotavirus was 62 adult males , 18 to 45 years old . The inoculation was oral . Unlike the previous dose-response empirical datasets , neither the Cumulative Dose model nor the Exponential model produce a good fit . The Beta-Poisson model ( ) was statistically a better fit than the Exponential model [20] . Both the Exponential and the Cumulative Dose model increase too rapidly in relation to the probability of infection of 1; i . e . these models cannot maintain a non-zero or non-one probability of infection for a dose range of several orders of magnitude . Conversely , the Beta-Poisson model does not suffer from this limitation since its convergence to 1 is slower , providing a wider range of variance ( Figure 6 and Table 3 ) . A possible explanation of the poor fit of the Cumulative Dose model is the high degree of acquired immunity to Rotavirus and the changing serotype profile circulating within populations [23] . Unlike the polio virus study , the rotavirus cohort consisting of adults ( 18–45 years old ) , is likely to have been exposed multiple times to various rotavirus serotypes [24] . Such heterogeneity in susceptibility flattens out dose response curves beyond what can be captured by exponential dose response models or this Cumulative Dose response model . In the previous subsections we fixed temporal exposure length , , to 1 hour , and assume that this is the time corresponding to the single shot inoculation , analogous to existing experimental dose-response trials . In this section , we present simulations for a range of different temporal exposure lengths , illustrating how longer times affect the dose response curve . The model is set to the parameters that provided an optimal fit for a temporal exposure length of . Figure 7 shows the dose-response curves for Poliovirus type 1 for different lengths of exposure for the estimated parameters used in Figure 4 to fit the experimental data for the condition Te = 1 . 0: . As the exposure length increases , the probability of infection decreases dramatically . Therefore , assuming that the unit of time is one hour , and this is the equivalent for a dose that is administered in a single shot , the probability of infection generated by the Cumulative Dose model for a dose of of 90 pathogens administered in one hour is 0 . 82 . If the dose were administered not in one hour , but uniformly over ten hours the probability of infection would be 0 . 18 . If the dose were administered over fifty hours the probability of infection would be reduced to 0 . 0001 . To obtain the same probability of infection for a ten hours inoculation period instead of one , we would require a dose of 139 pathogens instead of 90 . Because data on the impact of temporal patterns of inoculation are currently not available , a model with dose-time dependence such as ours is not identifiable [25]; i . e . , the model can be fit to existing single dose empirical data with many different parameters sets . For example , in Figure 8 we show model simulation results for Cryptosporidium parvum for two different parameter sets . Both parameters sets have a similar fit to the Cryptosporidium parvum dataset when ( mean square error using and is 3 . 5×10−3 and 9 . 7×10−3 respectively ) . For values of , however , the dose response relationships of the two parameter sets diverge . Parameter set is much less sensitive to exposure time than due its slower dynamics . Using parameter set R , pathogens proliferate faster , are being eliminated by each immune effector more quickly , are recruiting fewer immune effectors , and are eliminating immune effectors at a slower rate . On the other hand , using parameter set R , the natural rate of turnover of immune effectors is more rapid . We cannot argue at this point which is the most plausible configuration since identifiability cannot be resolved without data from dosing trials for different exposure lengths . In this section we relax the assumption that pathogens are inoculated at a fixed rate . We allow variation both in dose magnitude and length of exposure time , in order to capture a more realistic exposure scenario . The temporal pattern of inoculation of pathogens within a host depends both on the behavior of the host and the contamination of the environment the host interacts with . For instance , a susceptible host in a venue contaminated with influenza will be exposed to pathogens from air and fomites . However , the temporal patterns of exposure for these two modes of transmission are different . The host is likely to receive a small dose with every breath when breathing contaminated air . In fomite mediated transmission , however , the touching of a mucous membrane with contaminated fingers , for example , is likely to transmit a larger but less frequent dose . To illustrate this effect we devised an experiment where both the total inoculated dose and the exposure time length are fixed . The only parameter that varies is the number of inoculation events , , which ranges from 1 to the total dose . Consequently , once the number of inoculations events is determined , the dose inoculated in each event is and the rate at which inoculation occur is . Figure 9 shows the results of this experiment where the same parameter sets are used as in Figure 8 . The pathogen is Cryptosporidium parvum , and the same two different parameters sets , S and R , are used to inform the cumulative dose model . The total dose inoculated is set to and the temporal exposure length is set to Te = 120 . 0 hours . For both parameter sets S and R we observe the same behavior: infectivity decreases as the frequency or number of inoculations events increases . The temporal pattern more likely to be associated with fomite transmission ( low frequency and high dose , Figure 9 . B ) is more likely to produce infection than the patterns associated with airborne transmission ( high-frequency and low dose , Figure 9 . C ) . For parameter set R , the probability of infection if the dose is inoculated with a single exposure ( Figure 9 . A ) is 0 . 752 . The same dose inoculated over 4 events , where each event is one fourth of the total dose ( Figure 9 . B ) , reduces the probability of infection to 0 . 443 . In addition , if the dose is inoculated over 50 events ( Figure 9 . C ) the probability decreases to 0 . 111 . For parameter set S , the reduction of the infection probability is less pronounced: 0 . 740 , 0 . 676 and 0 . 601 for 1 , 4 and 50 inoculation events respectively . In previous sections we showed that longer temporal exposure lengths decrease infectivity due to the action of the immune system . In this section , we show that not only the duration of the exposure matters , but also the way in which pathogens arrive within that interval can decrease infectivity . These results suggest that risk assessments based on current dose-response data might be over-estimating risk of infection . An important corollary is that risk of infection for a given exposure dose may depend on the route of transmission based on their differences in the pattern of exposure .
We examined a dynamic mechanistic model where immune system effects generated dose response dependence on the timing of doses . The specific aspects of our model that generate these dose-timing effects are: 1 ) decreases in available immune effectors because they are being eliminated as they kill pathogens; and 2 ) increases in available immune effectors due to both pathogen dependent and independent recruitment . An additional mechanism resulting in decreases in available immune effectors that is not included in our model could be the time of immune effector engagement with pathogens in the killing process . The dose-timing effects we illustrate would be absent in a model where some effector like a T-cell instantaneously kills pathogens or pathogen generating cells , where no killing capacity is lost with each kill , and where effector dynamics are not otherwise altered by encounters with pathogens . Any such model , however , is highly unrealistic , and therefore we conclude that the dose-timing effects presented in our model could be important and warrant further study . Dose-timing effects have implications for microbial risk assessment , for infection transmission system modeling , and for the evolution of emerging pathogens . Considering a microbial risk assessment example , the implications of our findings suggest that exposure routes with different dose-timing dynamics could have different risks and therefore result in different clean up protocols for contamination events such as a norovirus outbreak or a Katrina-like disaster . Dose timing could , therefore , affect decisions on which venues to close or what the total dose that workers would be permitted to accrue during a cleanup operation . Considering modeling infection transmission , the standard approach is to define a contact and a transmission probability per contact while the physical route of transmission is ignored . Modeling the physical route of transmission is important when it is necessary to specify how much transmission is taking place in particular public venues and when specifying which control actions in these venues will reduce transmission . When different routes have different temporal exposure patterns , we demonstrate here that there is considerable potential for immune system effects to alter the ratio by which airborne transmitted and hand-fomite transmitted pathogens generate new infections . If we had data on infection risks under different dose-timing patterns , we could say more precisely how much difference in risk there might be from an airborne and a hand-fomite mediated pathogen . Unfortunately such data is lacking . The evolution of emerging infection implications derive from the route of transmission effects just discussed . When pathogens first jump species , they are likely to encounter strong innate immune responses to which they must evolve some escape strategy . That means very high transmission doses will be required to sustain transmission and that low dose exposure over longer times such as occurs with airborne transmission will be the most unlikely to be effective in transmitting infection . But , as escape from innate immune responses evolves , the balance could begin to favor airborne transmission which might be more effective in disseminating infection . We do not have enough dose timing data for any infection to evaluate either the microbial risk assessment implications , the infection transmission system implications , or the emerging infection evolution implications . Any data providing indications of the magnitude of dose-timing effects generated by any type of immunity to any agent would provide an important first step that would at least indicate what range of effects might be expected . Animal studies could compare the risks associated with a single instantaneously delivered dose with the same dose magnitude delivered over extended periods of time . Measurements of specific immune effector dynamics , such as interferon gamma [26] would improve our mechanistic understanding of a cumulative dose effect and indicate how to refine our models for different animal/pathogen systems . The issue of dose-response trial design is crucial for advancing both quantitative microbial risk assessment and analysis of population infection transmission systems . Due to the absence of a prior theoretical framework , there has been no motivation to conduct dosing trials that take multiple doses and multiple dosing times into account . Now that the potential effects of dose timing have been demonstrated and the practical significance of such measurements for microbial risk assessment and transmission system analyses is more evident , we hope to see such experiments . | We model the relationship between the temporal patterns of pathogen exposure and infection take off within people . Since different routes of transmission ( e . g . , airborne versus surface transfer routes ) may result in different temporal patterns of exposure , this model helps to better compare the risks of transmission from one person to another through these different routes . Previous models assumed that the risk of infection is the same whether pathogens are inoculated all at once or over one day . Our model , in contrast , captures how one pathogen affects the potential of immunity to keep concurrently or subsequently arriving particles from initiating an infection . Since the pattern of timing of airborne and surface spread pathogen arrivals differ , our model shows that each airborne pathogen could carry less risk than each surface transmitted pathogen . Unfortunately , data to fully fit our model are not currently available . Therefore new experiments will have to be conducted where doses are given across different temporal windows . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [
"mathematics",
"immunology/immune",
"response",
"immunology/innate",
"immunity",
"public",
"health",
"and",
"epidemiology/infectious",
"diseases",
"immunology/immunity",
"to",
"infections",
"infectious",
"diseases/epidemiology",
"and",
"control",
"of",
"infectious",
"diseases"
] | 2009 | The Effect of Ongoing Exposure Dynamics in Dose Response
Relationships |
MicroRNA ( miRNA ) –mediated gene regulation is of critical functional importance in animals and is thought to be largely constrained during evolution . However , little is known regarding evolutionary changes of the miRNA network and their role in human evolution . Here we show that a number of miRNA binding sites display high levels of population differentiation in humans and thus are likely targets of local adaptation . In a subset we demonstrate that allelic differences modulate miRNA regulation in mammalian cells , including an interaction between miR-155 and TYRP1 , an important melanosomal enzyme associated with human pigmentary differences . We identify alternate alleles of TYRP1 that induce or disrupt miR-155 regulation and demonstrate that these alleles are selected with different modes among human populations , causing a strong negative correlation between the frequency of miR-155 regulation of TYRP1 in human populations and their latitude of residence . We propose that local adaptation of microRNA regulation acts as a rheostat to optimize TYRP1 expression in response to differential UV radiation . Our findings illustrate the evolutionary plasticity of the microRNA regulatory network in recent human evolution .
microRNAs ( miRNAs ) are endogenous small RNAs that bind to their target mRNAs to post-transcriptionally repress protein production . They recognize their target mRNAs primarily through sequence complementarity between the seed region of a miRNA ( usually defined as the 2nd to the 7th or 8th nucleotide along a miRNA 5′ end ) and the binding sites on its target mRNAs [1] . A large number of human genes are known to be regulated by miRNAs , thus , miRNAs constitute a critical post-transcriptional regulatory network that plays vital roles in a broad range of biological processes [2]–[4] . Their functional importance is also consistent with the evolutionary conservation of miRNA-mediated regulations , as many miRNAs and their targets are conserved across species [5] , [6] and sequence variants that disrupt miRNA regulation are typically rare in humans and are often associated with human diseases [7]–[9] . However , human phenotypic evolution can be caused by changes in gene regulation , perhaps even more so than by changes in the proteins themselves [10]–[13] . Modulation of miRNA regulations is a possible path for such adaptive changes , but hitherto no solid evidence has been presented in favor of miRNA interactions playing an important role in human evolution . To address this issue , we first examined the degree to which local human adaptation is affected by changes in miRNA regulatory interactions , and then experimentally verified some of the identified interactions showing extreme population differentiation , including the regulation by miR-155 on TYRP1 to modulate human pigmentation phenotype .
We first aim to identify miRNA regulatory interactions that have been significantly differentiated among human populations , and then determine whether these differentiation events were driven by positive selection during human evolution . We mapped the ∼3 million HapMap Phase II single-nucleotide polymorphisms ( SNPs ) [14] onto predicted binding sites of known human miRNAs using TargetScanS [15] , and identified 2 , 217 bi-allelic SNPs that have one allele disrupting an intact binding site . We considered only those binding sites that have high-confidence scores ( see Materials and Methods , and Figure S1A ) . Population differentiation of each SNP was then quantified by FST , a commonly used statistical measure of genetic differentiation [16] . A SNP locus that has exlucively alternate alleles among populations receives an extreme FST value . We used FST computed by Barreiro et al . [17] for the 4 HapMap populations: Yoruba from Ibadan , Nigeria ( YRI ) , Japanese from Tokyo ( JPT ) , Han Chinese from Beijing ( CHB ) and Utah residents with northern and western European ancestry ( CEU ) . Our subsequent analysis revealed that many SNPs showing extreme population differentiation interfere with their predicted miRNA regulatory interactions ( Table S1 ) . Given the short evolutionary history of human population differentiation , this finding is in contrast with the common thought that miRNA-mediated regulation is strongly conserved [5] , [6] . Extreme population differentiation can be attributed to several possible factors , such as outliers from neutral drift , population structure , or positive selection for local adapation . Thus we sought to determine the potential sources accounting for the observed differentiation events . Recent population genetic analyses of human variation have shown that much of the recent local adaptation in humans may be caused by subtle changes in allele frequencies in many genes rather than strong changes in a few genes [18] , [19] . The effect of local ( population-specific ) selection can then be detected by comparing genome-wide pattern of FST . For example , Coop et al . showed that SNPs with high FST values in humans are enriched for genic SNPs in comparison to non-genic SNPs , and interpreted this as evidence of local selection targeting sites in genic regions [18] . Inspired by this work , we here ask whether a similar enrichment exists for miRNA target sites compared to the background distribution . Since the miRNA binding sites in this analyses were on the 3′ UTRs of human genes , we then collected a total of 23 , 030 3′ UTR SNPs to serve as a background control . We divided FST values into different bins ( Figure 1A ) , and computed the enrichment scores for SNPs in each bin . The enrichment score is the fraction of miRNA target-site SNPs in each bin , divided by the fraction of all 3′ UTR SNPs in the same bin . To determine the distribution of the enrichment score under the null hypothesis of no enrichment , we generated 1 , 000 data sets with 2 , 217 randomly sampled 3′UTR SNPs . Figure 1A clearly shows that loci with extreme FST values ( FST≥0 . 5 ) were significantly enriched for miRNA binding sites ( P<2 . 6×10−4; hypergeometric tail probability ) . Positive selection by definition acts on functional loci so these findings clearly indicate that miRNA binding sites with extreme population differentiation ( FST≥0 . 5 ) are likely targets of positive selection . We next focused on the polymorphic binding sites that have the strongest evidence of population differentiation ( FST≥0 . 5 ) ; however we also attempted to control , at least partially , for hitchhiking effects by excluding target-site SNPs linked to any annotated functional variant that lies 500 Kb upstream or downstream of the SNPs ( following the protocol of HapMap , see Materials and Methods ) . Using these filters , we identified 30 SNPs located in the putative miRNA binding sites of 26 genes , which showed very strong evidence of population differentiation ( Figure S1B , also see Tables S1 and S2 ) . In addition to using FST for cross-population comparison , we also examined these 30 loci for evidence of selection within individual populations , using several other aspects of the data including haplotype structure ( the integrated haplotype scores , iHS [20] ) and excessive high-frequency derived alleles ( Fay-Wu's H test [21] ) . These revealed that approximately half of the identified SNPs showed evidence of selection using either test statistic ( Table S1 ) . To further elucidate whether selection has been acting directly on the target site , rather than on linked sites , we assessed composite likelihood ratios ( CLR ) to identify the location of a selective sweep [22] . Figure S2 shows several examples where selection signals can be clearly localized around the polymorphic binding sites using this method . While we cannot exclude that the particular sites identified using these analyses show high FST values due to linkage with other SNPs that have not been annotated , or in a few cases are false positives with high degree of differentiation due solely to genetic drift , the procedures we have used here are designed to maximize the probability that the candidate sites identified here are targets of selection . Based on these studies we chose 7 candidate loci for further functional analyses ( Table S3 ) . To validate whether the 7 predicted candidate loci indeed display differentiated miRNA regulations , we made three variants of the 3′ UTR for each locus fused downstream of the firefly luciferase coding sequence using the pMIR-REPORTER vector ( Figure 1B ) . The first two variants carried either the ancestral or derived allele of a SNP within the putative miRNA target , while the third deleted the entire site . In addition , we employed the unrelated G3R 3′ UTR sequence , which is derived from the chicken versican G3 domain , as a control . Each reporter was then transfected into HEK293T cells either with control miRNA ( miR-CTL ) or a miRNA mimic that corresponds to the predicted miRNA regulator . Among the 7 putative interactions examined , we found that the predicted sites on SMNDC1 ( survival motor neuron domain containing 1 ) , SLC25A19 ( solute carrier family 25 , member 19 ) , and TYRP1 ( tyrosinase-related protein 1 ) showed significant miRNA-dependent inhibition of luciferase expression in an allele-specific manner ( Figure 1C and 1D and Figure 2 ) , while the remaining 4 genes showed no evidence of regulation ( Figure S3 ) . For example , the ancestral ‘A’ allele of SNP rs1050755 in SMNDC1 interfered with miR-329 regulation ( Figure 1C ) . This allele is the most common allele in East Asians ( CHB and JPT ) and CEU , but is rare in YRI , which almost exclusively possess the derived miR329-targeted ‘G’ allele ( derived allele frequency is 0 . 98 ) . In the same vein , the ancestral allele ‘C’ of SNP rs7198 mediated allele-specific regulation of SLC25A19 by miR-122 , while its derived ‘G’ allele prevented regulation ( Figure 1D ) . Notably , the derived allele reaches a frequency around 90% in East Asian populations ( CHB and JPT ) but is rare in YRI ( with derived allele frequency 0 . 02 ) . The third validated interaction , between miR-155 and two linked SNPs mapped to the 3′UTR of TYRP1 was subject to extensive analysis ( Figure 2 and discussed below ) . Of note , given the difficulty in predicting miRNA targets and a lack of consensus among miRNA target prediction algorithms [23] , it is encouraging that we could validate 3 of the 7 in silico predicted interactions . Regardless , these results verify that the highly differentiated SNPs in several cases have a direct impact on miRNA-mediated regulation . Worthy of note , for two of the three validated sites ( on SLC25A19 and TYRP1 ) , the signal of positive selection can be localized directly to the miRNA binding sites using the CLR method ( Figure 1E ) . The alteration , rs7198 on SLC25A19 disrupted miR-122 regulation and is affected by the selective sweep , almost fixing the derived allele in East Asians with a frequency of ∼90% . This is further supported by analyses using other methods including Fay and Wu's H ( Figure 1F ) , where a sharp reduction in H suggests an excess of derived alleles fixed by the sweep around this locus . These signals , however , were absent in YRI and CEU ( Figure S4 ) , particularly in YRI where the derived allele is very rare . The second verified miRNA regulation with extreme population differentiation and strong evidence of selection acting directly on the miRNA binding sites , TYRP1 , is more interesting since it is associated with an obvious population differentiated phenotype – pigmentation . TYRP1 is an enzyme specifically expressed in melanocytes , which promotes melanin production and regulates pigmentation in skin , eyes and hair [24]–[26] . Mutations in TYRP1 can cause oculocutaneous albinism type 3 ( OCA3 ) [27] , [28] . Furthermore , TYRP1 is important for adjusting skin reflectance to protect against excessive UV exposure [29] and recent genome-wide association studies have consistently found this gene to be associated with differentiated pigmentation among populations [30]–[32] . Lastly , several studies have suggested that this gene has been under strong positive selection relating to adaptation to local environments [33]–[35]; however no causal variants have been identified . In our study , we discovered that TYRP1 harbors 2 common SNP variants , rs683 and rs910 that reside in two putative miR-155 binding sites ( Figure 2A ) . These two SNPs are nearly fixed for the derived allele in the African and Asian populations ( YRI , CHB and JPT ) , but remain polymorphic in the European population ( CEU ) . Notably rs683 is reported to be associated with difference in iris color among Europeans [30] . To investigate this link we next sought to systematically validate the role of these SNPs in population-specific miRNA regulation of TYRP1 . miR-155 is very narrowly expressed among human tissues [36] , [37] . Interestingly it is reported to be expressed in melanocytes [38] , suggesting it may physically interact with TYRP1 . Notably miR-155 is also an oncomir [38] , [39] , and is involved in cell signaling and dendritic development [40]–[43] . The 3′ UTR of TYRP1 contains three putative miR-155 binding sites , among which two are polymorphic in the HapMap data analyzed here . As seen in Figure 2A , site I is a non-canonical site mediating a 6-mer match to the miRNA seed and is located immediately downstream of the stop codon , while sites II and III mediate canonical pairing with the intact seed region of miR-155 . The two SNPs , rs683 and rs910 , reside within site II and site III respectively , with the derived alleles forming intact miRNA sites in CHB , JPT and YRI . In contrast , two thirds of CEU individuals carry the ancestral alleles that disrupt miRNA-target interaction . Due to their physical proximity , the two SNPs are tightly linked with D' = 1 and LOD = 24 . 44 , indicating their co-presence ( or co-absence ) in CEU individuals ( Figure 2A ) . To test the function of these putative miR-155 targets , 7 variants of the TYRP1 3′UTR harboring combinations of ancestral , derived and deleted miR-155 target elements were constructed in pMIR-REPORTER ( Figure 2B ) . Transfection into HEK293T cells either alone or in the presence of increasing amounts of synthesized miR-155 revealed that the 3′ UTR harboring the derived alleles ( construct A , Figure 2C ) was substantially suppressed by miR-155 . Furthermore , a mutant in which all three putative miR-155 targets were deleted ( construct C , Figure 2C ) showed minimal repression that was comparable to the G3R control ( G3R , Figure 2C ) . In contrast , when we analyzed the ancestral alleles ( construct B , Figure 2C ) , suppression was compromised , indicating that the sequence variations in sites II and III interfered with miR-155-dependent regulation of TYRP1 ( Figure 2C ) . The derived TYRP1 3′ UTR ( construct A ) is thus a target of miR-155 . Analysis of the ancestral alleles revealed substantial suppression when compared to a mutant in which all three miR-155 targets were deleted ( construct C , Figure 2C ) . This suggested that the fixed site I might be functional . Therefore , to determine the relative contributions of the three sites to miR-155 regulation of TYRP1 , we next tested various combinations of site deletants ( Figure 2B ) . This revealed that site II mediated the strongest suppression ( Figure 2D , curve E ) , while site I was weaker ( Figure 2D , curve D ) and site III was the weakest ( Figure 2D , curve F ) . As site III is linked with site II ( Figure 2A ) , we subsequently analyzed alternate site II alleles in isolation , since the strongest suppression mediated by this site might serve to direct natural selection on this locus . We observed that interruption of the site by the ancestral allele ( construct G ) strongly blocked miR-155 suppression ( Figure 2E ) . As TYRP1 is best known for its role in regulating human skin pigmentation , we next tested allele-specific regulation of the endogenous gene by miR-155 using a skin-derived cell line . SK-MEL-19 cells express endogenous TYRP1 , while this gene is not expressed in many other melanoma cell lines [44] . We examined the TYRP1 alleles in SK-MEL-19 cells and found it to be heterozygous for rs683 ( Figure 3A ) . Next we transfected these cells with either miR-CTL or increasing amounts of miR-155 , and monitored TYRP1 protein levels by immunoblotting . This revealed that endogenous TYRP1 protein level decreased with increasing miR-155 concentration ( Figure 3B ) , consistent with our analysis of the heterologous luciferase reporter assay . As miRNA can suppress protein production either through destabilization , or through translational inhibition , we next performed quantitative RT-PCR ( qPCR ) to examine TYRP1 mRNA levels , which were also reduced upon miR-155 transfection ( Figure 3C ) . Finally , to demonstrate that miR-155 preferentially targets the derived allele , we employed TaqMan SNP qPCR to detect the abundance of TYRP1 transcripts that carry either the ancestral or the derived alleles of rs683 . Transfection of increasing amounts of miR-155 led to a modest reduction in the ancestral allele that contrasted the much stronger suppression observed in the derived allele ( Figure 3D ) . Taken together , these studies establish that the derived alleles of rs683 and rs910 , which are almost fixed in YRI , CHB and JPT populations ( Figure 2A ) , introduce two additional miR-155 targets that serve to enhance miR-155-mediated suppression of TYRP1 . In contrast , these alleles segregate at a frequency of only approximately 1/3 in CEU , with 2/3 of the population carrying the alternate ancestral allele that interferes with miR-155-mediated suppression of TYRP1 expression . Previous studies suggested that TYRP1 has been under selection in different populations [33]–[35] , but none of the causal alleles were identified . We next investigated in more detail the pattern of selection that has driven miRNA site turnover in TYRP1 between human populations . Since our analysis of the selection signature on TYRP1 was based on inter-population comparison ( FST ) , we next tracked the local selection within individual populations . As shown in Figure 2A , the derived alleles of rs683 ( and also the linked site rs910 ) are almost fixed in YRI and East Asians ( CHB+JPT ) , and the CLR test [22] revealed a selection signature around this locus ( Figure 4A for East Asians , Figure S5A for YRI ) , with the signal peaking around the binding sites ( the dotted line ) . Consistent with the CLR test statistic , Fay-Wu's H statistic correspondingly showed a sharp reduction around the region of interest ( Figure 4B for East Asian , Figure S5B for YRI ) . These signals however are absent in CEU where the derived allele is in the minor form ( Figure S5C–S5D ) . We also did extensive analyses to explore the possibility of linkage disequilibrium ( LD ) between rs683 and rs910 and other known functional SNPs in the region , but could not detect any high-LD SNPs with annotated functions ( see Materials and Methods and also Figure S6 ) . Therefore it is most likely that the two miRNA binding sites mediated by rs683 and rs910 were direct targets of positive selection in YRI , CHB and JPT . Given that it is the derived states of the two SNPs that maintain miR-155 regulation ( Figure 2A ) , positive selection , which increased the derived allele frequencies of the two de novo binding sites on TYRP1 in YRI , CHB and JPT , likely reflects a requirement in these populations to induce miR-155 suppression on TYRP1 . In CEU the major allele is the ancestral form , and in this population there is no evidence of positive selection affecting the derived allele ( Figure S5C–S5D ) . However , we found high extended haplotype homozygosity ( EHH ) [45] for the ancestral alleles of rs683 and rs910 ( Figure 4C and Figure S7A ) , which is a sign of recent selection acting to expand the ancestral alleles in CEU . This trend is absent in other populations ( Figure S7B–S7C ) . Further , the integrated haplotype scores ( iHS ) for the ancestral alleles in CEU are all above 3 , substantially higher than the typical threshold of 2 used in humans for detecting loci subject to positive selection [20] . Since there are no other known functional SNPs in high LD with these SNPs in the HapMap data ( see Figure S6 ) , or in the more comprehensive set of SNPs reported in dbSNP 132 [46] , which includes data from the 1000 Genomes Project [47] , the observed increase in haplotype homozygosity in CEU is thus most likely explained by selection directly expanding the ancestral allele of rs683 , which compromises miR-155 regulation on TYRP1 . Taken together , these results suggest that in YRI , and particularly in CHB and JPT , the derived alleles of rs683 and rs910 have been a target of positive selection ( Figure 4A and 4B , and also Figure S5A and S5B ) , whereas in CEU , haplotypes carrying the ancestral alleles were recently selected for , leading to an increase in haplotype homozygosity among haplotypes carrying the ancestral allele in this population ( Figure 4C and Figure S7A ) . Due to the significant role of TYRP1 in modulating human pigmentation [26]–[29] , a possible adaptive explanation is that the miRNA binding sites on TYRP1 have been targeted by population-specific selection in relation to exposure to sun light ( UV irradiation ) . We investigated this hypothesis further by extending our analysis from the 4 HapMap populations to 37 representative indigenous populations genotyped in the Human Genome Diversity Project ( HGDP ) , spanning 650 , 000 common SNPs [48] . The SNP rs683 at site II is also genotyped in HGDP , which co-segregates with rs910 at site III ( Figure 2A ) . For each population , we correlated the absolute latitude where the population resides with the derived allele frequency of rs683 ( Figure 4D ) , and found a strong negative correlation ( Pearson's R = −0 . 63 ) . As the values for different populations might be correlated due to miration history , we cannot apply standard statistical methods to test whether the correlation is significant . However , we note that the correlation between latitude and allele frequency in rs683 is among the 1% most extreme of such correlations in the genome . This is also true if we restrict ourselves to analyses of SNPs with FST≥0 . 5 ( Figure S8A–S8B ) . Thus , the closer to the Equator that a population resides , the higher the frequency of the derived allele . There are several populations that show deviations from this trend , in particular the pygmy populations in Africa , which show less evidence of selection than the East Asian or European populations . TYRP1 has been suggested to be a target of positive selection , and the mode of selection on this gene is thought to be complicated [33]–[35] . Our analyses now reveal that positive selection has driven population differentiation of miR-155 regulation on this gene , providing new insights into the causes of the observed selection signatures ( Figure 4 ) . Indeed , TYRP1 has a well-established role in regulating skin pigmentation , and its expression is ∼2 . 6-folder higher in Africans than in Europeans [29] . Thus the presence of selection fixing miRNA binding sites on the highly expressed TYRP1 in Africans is a good example of incoherent regulation , i . e . when a miRNA represses a target gene in the direction opposing the overall outcome of all the other regulatory processes ( e . g . by transcription factors ) [49]–[51] ( not to be confused with the incoherent control or incoherent feed-forward loop in describing generic regulatory networks ) . Such a network architecture is important in maintaining target protein homeostasis and in fine-tuning and buffering target protein expression . Indeed , the amount of skin pigmentation is thought to be balanced between two conflicting and UV-dependent physiological needs , the production of vitamin D and folate [52] . Hyper-pigmentation can cause vitamin D deficiency , while hypo-pigmentation can cause folate deficiency , both being tightly associated with human reproductive success . Thus , pigmentation genes are likely to be highly regulated , and gain of additional miRNA binding sites on the highly expressed TYRP1 in Africans ( in low-latitude regions ) might ensure proper expression of this gene by dampening potential fluctuations that may shift its expression away from the optimal level , conferring unfavorable pigmentation phenotypes ( Figure 4E ) . On the other hand , in high latitudes with low UV exposure , light pigmentation and low TYRP1 expression is strongly favored . In these areas , recent selection that expanded the ancestral allele to disrupt miR-155 repression might suggest a physiological need to remove the excessive miRNA regulation on the already lowly expressed TYRP1 , which otherwise would cause hypo-pigmentation ( Figure 4E ) . Moreover , disrupting additional miRNA binding sites on this gene confers rapid response to external stimulus , and indeed analysis of TYRP1 expression by solar irradiation in a variety of populations including African and European revealed that only Europeans displayed significant induction of TYRP1 upon chronic photoexposure [29] . Therefore this mechanism might facilitate rapid adaptation to environment with elevated photoexposure . Taken together , our results reveal that the regulatory interaction between miR-155 and TYRP1 was highly plastic during human evolution; this may serve as a physiological rheostat to optimize the expression of TYRP1 to distinctively advantageous level in different populations , in response to differential UV radiation along the latitudes of their residence .
King and Wilson first proposed that most human phenotypic evolution may be due to changes in gene regulation [10] . This notion has been supported by a number of studies showing that a considerable proportion of the genetic variation underlying phenotypic human variation and human-chimpanzee differences may lie outside protein-coding regions [11]–[13] . However , genetic changes at the post-transcriptional level ( e . g . regulation by miRNAs ) have received little attention , and previous studies have not established a clear functional effect of alleles predicted to be under selection [7] , [8] . Our study now revealed that positive selection can drive population differentiation of human miRNA regulation , suggesting that miRNA regulation could be highly evolutionarily plastic , and may contribute to human evolution . We also found that a majority of the identified sites are in non-conserved elements revealed by genomic comparison across 17 vertebrates ( quantified by phastCons score [53] , which varies between 0 and 1 ) . For example , for the validated sites in this study , their highest phastCons scores are only around 0 . 1 , suggesting genetic novelty may arise from elements that are under relaxed selective pressure . Of note , we also scanned the known miRNA loci in our analysis , but did not find any miRNA loccus that has elevated differentiation among populations . This observation indicates that binding sites turnover may be a more prevalent mechanism in modulating miRNA regulation than changing miRNAs themselves . This notion is supported by a recent study which re-sequenced known miRNA loci among human populations and found an absolute lack of sequence diversity within the miRNA seed regions [54] . There are likely many more changes in miRNA interactions that have contributed to human adaptation than reported here . First , we only report signals relating to increased levels of population differentiation , which most likely only reveal a small fraction of the selection that has acted during human evolution since the divergence with chimpanzees . We also used a very stringent cutoff for FST ( ≥0 . 5 ) in this study since we aimed to find the most extreme cases of population differentiation . Future comparative studies aimed at miRNA regulatory sites may reveal many more examples of rewired miRNA interactions . Second , our target prediction was based on TargetScanS , which uses seed matches as the principal mechanism for target recognition by miRNAs . However , other mechanisms might also exist , such as sites with central pairing [55] , and it is possible that natural selection might act on sites lacking the canonical seed matches . Third , in this study , we scanned 3′UTRs of human genes for putative miRNA binding sites , however , increasing evidence has shown miRNA might also target coding regions or 5′UTRs , although repression strength is more marginal [56] . Future studies will be required to validate other putative sites identified in our study and to elucidate the underlying evolutionary significance of the selection signature .
SNP annotations were based on a previous study [17] which calculated Fst values for ∼3 million HapMap Phase II SNPs . When controlling for potential hitchhiking effects , we followed the protocol used in HapMap database and computed the pairwise linkage disequilibrium ( LD , quantified by R2 ) between the SNPs in question and its flanking SNPs within 500 Kb downstream and upstream . Any SNPs having R2≥0 . 5 with annotated functional sites within this distance were excluded from the analysis . The function annotation of SNPs was retrieved from dbSNP 130 queried from UCSC Table Browser . Following previous protocol [20] , we assigned human SNPs with ancestral alleles based on the chimpanzee reference genome ( queried from UCSC Table Browser ) . For the 30 SNPs of particular interest , we also confirmed its ancestral allele by comparing with the reference genomes of orangutan and rhesus macaque ( Figure S1B ) . Allele frequencies of rs683 in world populations were collected from the Human Genome Diversity Project [48] , and were extracted from the UCSC Table Browser . For each population genotyped , we extracted its absolute latitude ( the absolute value of the latitude ) , which indicates the angle of a location from the Equator rather than relative north and south . Among the genotyped populations , we used Han to represent ∼92% of Chinese population , and excluded the Chinese minorities from our analysis due to their complicated ethnohistorical characteristics and migration histories . We also excluded Yakut as they are very recent migrants approximately 1 , 000 years ago , with an effective female population size of only 150 individuals [57] . This is because population migration might distort our analysis of long-term selection in particular population residence . The iHS values and Fay and Wu's H for HapMap populations were extracted from Haplotter ( http://haplotter . uchicago . edu/ ) . CLR test was performed using SweepFinder [22] by setting the background site frequency spectrum estimated from all SNPs across the genome . SNPs from HapMap ( rel . 27 ) were subject to this analysis . To explore the possibility that rs683 and rs910 changed frequency due to hitchhiking with some other functional variants on this gene , we analyzed all the known HapMap SNPs on TYRP1 in YRI and CEU , and computed their linkage disequilibrium ( LD ) R2 with rs683 using PGEToolbox [58] ( Figuere S8 ) . We also considered SNPs in the 5-Kb upstream region of TYRP1 . Most SNPs that showed significant LD with rs683 are intronic and do not overlap with any annotated splice sites , while two loci in the 5′ upstream region showed only weak LD . Although one missense SNP was found , it was not linked with rs683 and displayed an extremely low LD . In contrast , two SNPs rs2762464 and rs1063380 , located in the 3′UTR of TYRP1 ( Figure S6 ) , were within a strong linkage disequilibrium region of rs683 and rs910 , consistent with their close physical proximity ( <300 bp away ) . However , neither has any known mechanistic association with TYRP1 and none of the SNPs mediates interaction with known miRNAs . Similarly in CHB and JPT , we did not find any known functional variants on TYRP1 in strong LD with rs683 . We also expanded the analysis from HapMap SNPs to dbSNP132 by querying the UCSC Table Browser , which includes SNPs genotyped in the 1000 Genomes Project [47] , and did not find any annotated functional variants in high LD with rs683 . We first extracted all SNPs annotated to be in 3′ UTR of human genes , which were annotated by HapMap ( rel . 27 for all populations ) , and all the SNPs are polarized to the plus strand . These 3′UTR SNPs were then mapped onto the 3′UTR sequences of RefSeq transcripts ( downloaded from UCSC Table Browser as of July , 2010 ) , and we retained the longest transcripts when multiple sequences are annotated under the same transcript . Sequences with inconsistent annotations were discarded from our analysis . With this mapping procedure , we then generated a set of polymorphic 3′UTR segments , which is a 15 nucleotide window centered at the SNP position . Therefore for each SNP , sequences within the window will present twice each carrying the alternate alleles . TargetScanS [15] was then implemented to scan the collection of the polymorphic 3′UTR segments , and the predicted sites were then identified , which encompass a SNP , one of whose alleles does not interact with any miRNA while the other is miRNA-interacting . Prediction confidence was determined by context scores assigned the prediction program and we considered confident sites if their context scores no more than −0 . 2 ( Figure S1A ) . We considered 545 miRNA families deposited in TargetScanS ( Table S4 ) . For experimental validation , we further scanned the putative miRNA sites in fine solution by allowing a 6-mer match . All the constructs for this study were derived from the pMIR-REPORTER ( Ambion ) . Human 3′UTR sequences in this study were amplified from genomic DNA of HEK293T cells using PCR . The PCR products were subcloned into pMIR-REPORTER vector . Overlapping PCR and QUICKCHANGE II XL Site-Directed Mutagenesis Kits ( Agilent ) were used to generate the mutants of 3′UTRs containing different SNPs and deletions of miRNA target sites . pMIR-REPORTER β-galactosidase vector was used as the transfection control . G3R is a gift from Dr . Burton Yang' lab at the University of Toronto , which has the coding sequence of the chicken versican G3 domain in the pMIR-REPORTER vector , and was used as a negative control . All the synthesized miRNA mimics and the mimic negative control were purchased from Dharmacon . miR-CTL in this study is a negative control with sequence from a C . elegans miRNA cel-miR-67 , which has minimal sequence identity with miRNAs in human , mouse and rat ( by Dharmacon ) . HEK293T cells were grown in high glucose DMEM containing 10% FBS ( Thermo ) . They were transiently transfected with DNA constructs and miRNA mimics at 40% confluency in 24-well plates using the calcium-phosphate precipitation method . Cells were lysed 48 hours after transfection , and the activities of firefly luciferase and β-galactosidase of total cell lysates were determined using the Firefly Luciferase Assay System ( Promega ) and the β-gal assay previously described , respectively . To obtain the relative activity , the activity of the firefly luciferase was first normalized to the activity of β-galactosidase to obtain normalized firefly luciferase activity ( nFFLuc ) , and the data were then determined using the following formula: Relative Activity = log2 ( nFFLucmiR-155/nFFLucmiR-CTL ) , where nFFLucmiR-155 or nFFLucmiR-CTL is the nFFLuc in the presence of mir-155 mimics or mimic negative control . The concentration of miR-155 was increased as indicated in the figures ( Figure 1C and 1D , Figure 2C–2E , and Figure 3B–3D ) , with the concentration of miR-CTL being 2 nM . If Relative Activity was positive , we manually set it to be 0 . All the positive Relative Activity are no more than 0 . 13 . SK-MEL-19 cells were grown in RPMI 1640 ( GIBCO ) containing 10% FBS and antibiotics . For genotyping , the lysates of SK-MEL-19 cells were subjected to PCR . The forward primer starts from 116 bps upstream of rs683 whereas the reverse primer starts from 118 bps downstream of rs683 . The PCR products were cleaned and sent to DNA sequencing using the forward primer . The sequencing spectra were processed and analyzed using MacVector . SK-MEL-19 cells were transiently transfected with miRNA mimics following the instructions of the RNAi MAX transfection kit ( Invitrogen ) . For immunoblotting analysis , cells lysates were collected 40 hours after transfection . The antibodies used for immunoblotting analysis are: TYRP1 ( sc-10443 , Santa Cruz ) and GAPDH ( G9545 , Sigma ) . The ΔCT values were obtained by comparing the amplification of the target cDNA to that of HPRT . The TapMan SNP qPCR kit for rs683 ( C_3119206_10 , AB Biosystems ) was used to analyze the expression of the TYRP1 transcripts carrying its alternate alleles . This kit included two probes , one conjugated with VIC fluorescence dye to monitor transcripts carrying the ancestral allele and the other conjugated with FAM fluorescence dye to detect the derived allele . To evaluate the cross-hybridization between the probes , a pilot qPCR analysis was performed for the constructs A and B in Fig . 2B , each carrying the ancestral and derived alleles , respectively ( Figure S9 ) . A total of 8 samples by mixing the construct A and B were prepared as the following ratios of A to B: 1∶0 , 1∶1 , 1∶2 , 1∶4 , 4∶1 , 2∶1 , 0∶1 . The total DNA concentrations of the 8 samples were constant with 0 . 1 ug/ul . The same amount of DNA were then taken from these 8 samples individually to dilute 1000 times in water and these 8 diluted samples were subject to TaqMan SNP qPCR analysis according to the instructions of the kit . Expected Relative Fraction refers to the fraction of the construct in the mixed samples relative to the construct in the sample without mixing the other construct . Observed Relative Fraction was obtained using qPCR ΔCt of the construct in the mixed sample relative to the ΔCt of the construct in the sample without mixing the other construct . Using Observed and Expected Relative Fractions , two regression lines were plotted for the FAM and VIC signals to determine the effects of cross-hybridization between the probes . We then used these probes to detect the relative expression of TYRP1 transcripts carrying different alleles in the SK-MEL-19 cells , and probe intensities after miR-155 transfection were obtained after normalizing to the data points by transfecting miR-CTL . | MicroRNAs ( miRNAs ) are endogenous small RNAs that bind to their target mRNAs to post-transcriptionally repress protein production . miRNA–mediated gene regulation is usually considered to be strongly conserved among and within species , and thus alteration of such regulations is usually considered as detrimental . However , it is likely that evolutionary divergence of miRNA regulation may actually be selectively advantageous and could even serve as a genetic reservoir for innovation and adaptation . Towards this goal , we identified a number of polymorphic miRNA binding sites that display extreme population differentiation and show evidence of positive selection . We experimentally validated 3 regulations , including a regulation by miR-155 on TYRP1 , a melanosomal enzyme associated with human pigmentation . We found that the two alternate alleles on the 3′ UTR of TYRP1 , either inducing or disrupting repression by miR-155 , are under opposite selections among human populations . This results in a strong negative correlation between the degree of fixation of miR-155–mediated repression of TYRP1 in a population and the population's latitude of residence . These observations collectively suggest miR-155 acts a rheostat to optimize TYRP1 expression for local adaptation to differential UV radiation along the latitudes . Our findings demonstrate the plasticity of miRNA regulation in recent human evolution . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"systems",
"biology",
"genomics",
"biology",
"computational",
"biology",
"evolutionary",
"biology",
"genetics",
"and",
"genomics"
] | 2012 | Evidence for Positive Selection on a Number of MicroRNA Regulatory Interactions during Recent Human Evolution |
Telomere integrity in budding yeast depends on the CST ( Cdc13-Stn1-Ten1 ) and shelterin-like ( Rap1-Rif1-Rif2 ) complexes , which are thought to act independently from each other . Here we show that a specific functional interaction indeed exists among components of the two complexes . In particular , unlike RIF2 deletion , the lack of Rif1 is lethal for stn1ΔC cells and causes a dramatic reduction in viability of cdc13-1 and cdc13-5 mutants . This synthetic interaction between Rif1 and the CST complex occurs independently of rif1Δ-induced alterations in telomere length . Both cdc13-1 rif1Δ and cdc13-5 rif1Δ cells display very high amounts of telomeric single-stranded DNA and DNA damage checkpoint activation , indicating that severe defects in telomere integrity cause their loss of viability . In agreement with this hypothesis , both DNA damage checkpoint activation and lethality in cdc13 rif1Δ cells are partially counteracted by the lack of the Exo1 nuclease , which is involved in telomeric single-stranded DNA generation . The functional interaction between Rif1 and the CST complex is specific , because RIF1 deletion does not enhance checkpoint activation in case of CST-independent telomere capping deficiencies , such as those caused by the absence of Yku or telomerase . Thus , these data highlight a novel role for Rif1 in assisting the essential telomere protection function of the CST complex .
Telomeres , the specialized nucleoprotein complexes at the ends of eukaryotic chromosomes , are essential for genome integrity . They protects chromosome ends from fusions , DNA degradation and recognition as DNA double-strand breaks ( DSBs ) that would otherwise lead to chromosome instability and cell death ( reviewed in [1] ) . Telomeric DNA in the budding yeast Saccharomyces cerevisiae , as well as in nearly all other eukaryotes examined to date , comprise short TG-rich repeated sequences ending in a short single-stranded 3′ overhang ( G tail ) that corresponds to the strand bearing the TG-rich repeats . The addition of telomeric repeats depends on the action of telomerase , a specialized reverse transcriptase that extends the TG-rich strand of chromosome ends . Recruitment/activation of this enzyme requires the Cdc13 protein that binds to the telomeric TG-rich single-stranded DNA ( ssDNA ) [2]–[6] . The direct interaction between Cdc13 and the Est1 regulatory subunit of telomerase is essential for telomerase recruitment , and it is disrupted by the cdc13-2 mutation that leads to gradual telomere erosion and accompanying senescence [2] , [4] , [7] . The average length of S . cerevisiae telomeric 3′ overhangs is 12–14 nucleotides , although it can increase to ∼50 nucleotides during the late S/G2 phase of the cell cycle [8]–[10] . While single-stranded telomeric G-tails can arise after removal of the last RNA primer during lagging-strand replication , the blunt ends of the leading-strand telomere must be converted into 3′ overhangs by resection of the 5′ strand . This 5′ to 3′ nucleolytic degradation involves several proteins , such as the MRX complex , the nucleases Exo1 and Dna2 and the helicase Sgs1 [10] , [11] . Cyclin-dependent kinase activity ( Cdk1 in S . cerevisiae ) is also required for generation of the extended single-stranded overhangs in late S phase [12] , [13] . As Cdk1 activity is low in G1 , telomere resection can occur only during S/G2 [8] , coinciding with the time frame in which G-tails are lengthened and can serve to recruit telomerase . Keeping the G tail in check is crucial to ensure telomere stability , and studies in budding yeast have shown that Cdc13 prevents inappropriate generation of ssDNA at telomeric ends [2] , [14] , [15] . This essential capping function depends on Cdc13 interaction with the Stn1 and Ten1 proteins to form the so-called CST ( Cdc13-Stn1-Ten1 ) complex . This complex binds to telomeric ssDNA repeats and exhibits structural similarities with the heterotrimeric ssDNA binding complex Replication protein A ( RPA ) [16] , suggesting that CST is a telomere-specific version of RPA . Loss of Cdc13 function through either the cdc13-1 temperature sensitive allele or the cdc13-td conditional degron allele results in telomere C-strand degradation , leading to activation of the DNA damage checkpoint [13] , [14] , [17] , [18] . Similarly , temperature sensitive mutations in either STN1 or TEN1 genes cause telomere degradation and checkpoint-mediated cell cycle arrest [19]–[21] . Interestingly , Stn1 interacts with Pol12 [22] , a subunit of the DNA polymerase α ( polα ) -primase complex with putative regulatory functions , while Cdc13 interacts with the polα catalytic subunit of the same complex [7] , suggesting that CST function might be tightly coupled to the priming of telomeric C strand synthesis . In any case , it is so far unknown whether the excess of telomeric ssDNA in cst mutants arises because the CST complex prevents the access of nuclease/helicase activities to telomeric ends and/or because it promotes polα-primase-dependent C strand synthesis . In addition to the capping function , a role for the CST complex in repressing telomerase activity has been unveiled by the identification of cdc13 , stn1 and ten1 alleles with increased telomere length [2] , [21] , [23] , [24] . The repressing effect of Cdc13 appears to operate through an interaction between this protein and the C-terminal domain of Stn1 [25] , [26] , which has been proposed to negatively regulate telomerase by competing with Est1 for binding to Cdc13 [4] , [24] . A second pathway involved in maintaining the identity of S . cerevisiae telomeres relies on a complex formed by the Rap1 , Rif1 and Rif2 proteins . Although only Rap1 is the only shelterin subunit conserved in budding yeast , the Rap1-Rif1-Rif2 complex functionally recapitulates the shelterin complex acting at mammalian telomeres ( reviewed in [27] ) . Rap1 is known to recruit its interacting partners Rif1 and Rif2 to telomeric double-stranded DNA via its C-terminal domain [28]–[30] . This complex negatively regulates telomere length , as the lack of either Rif1 or Rif2 causes telomere lengthening , which is dramatically increased when both proteins are absent [30] . The finding that telomere length in rif1Δ rif2Δ double mutant is similar to that observed in RAP1 C-terminus deletion mutants [30] suggests that Rap1-dependent telomerase inhibition is predominantly mediated by the Rif proteins . However , Rif proteins have been shown to regulate telomere length even when the Rap1 C-terminus is absent [31] , suggesting that they can be brought to telomeres independently of Rap1 . In addition to negatively regulate telomere length , Rap1 and Rif2 inhibit both nucleolytic processing and non homologous end joining ( NHEJ ) at telomeres [32]–[34] . Telomeric ssDNA generation in both rif2Δ and rap1ΔC cells requires the MRX complex [33] , and the finding that MRX association at telomeres is enhanced in rif2Δ and rap1ΔC cells [33] , [35] suggests that Rap1 and Rif2 likely prevent MRX action by inhibiting MRX recruitment onto telomeric ends . Interestingly , the checkpoint response is not elicited after inactivation of Rap1 or Rif2 , suggesting that either the accumulated telomeric ssDNA is insufficient for triggering checkpoint activation or this ssDNA is still covered by Cdc13 , which can inhibit the association of the checkpoint kinase Mec1 to telomeres [36] . Notably , Rif1 is not involved in preventing telomeric fusions by NHEJ [32] and its lack causes only a slight increase in ssDNA generation at a de novo telomere [33] . These findings , together with the observation that Rif1 prevents telomerase action independently of Rif2 , indicate that Rif1 and Rif2 play different functions at telomeres . As both CST and the shelterin-like complex contribute to telomere protection , we asked whether and how these two capping complexes are functionally connected . We found that the viability of cells with defective CST complex requires Rif1 , but not Rif2 . In fact , RIF1 deletion increases the temperature sensitivity of cdc13-1 cells and impairs viability of cdc13-5 cells at any temperature . Furthermore , the rif1Δ and stn1ΔC alleles are synthetically lethal . By contrast , the lack of Rif2 has no effects in the presence of the same cdc13 and stn1 alleles . We also show that cdc13-1 rif1Δ and cdc13-5 rif1Δ cells accumulate telomeric ssDNA that causes hyperactivation of the DNA damage checkpoint , indicating that loss of Rif1 exacerbates telomere integrity defects in cdc13 mutants . By contrast , deletion of RIF1 does not enhance either cell lethality or checkpoint activation in yku70Δ or est2Δ telomere capping mutants . Thus , Rif1 is required for cell viability specifically when CST activity is reduced , highlighting a functional link between Rif1 and CST .
Yeast cells harbouring the cdc13-1 temperature-sensitive allele of the gene encoding the essential telomeric protein Cdc13 are viable at permissive temperature ( 20–25°C ) , but die at restrictive temperature ( 26–37°C ) , likely due to accumulation of ssDNA at telomeres caused by the loss of Cdc13 capping functions [14] . As also the shelterin-like complex contributes to the maintenance of telomere integrity , we investigated its possible functional connections with Cdc13 by disabling either Rif1 or Rif2 in cdc13-1 cells . Deletion of RIF2 did not affect cdc13-1 cell viability in YEPD medium at any tested temperature ( Figure 1A ) . By contrast , cdc13-1 rif1Δ cells showed a maximum permissive temperature for growth of 20°C and were unable to grow at 25°C , where cdc13-1 single mutant cells could grow at almost wild type rate ( Figure 1A ) . The enhanced temperature-sensitivity of cdc13-1 rif1Δ cells was due to the lack of RIF1 , because the presence of wild type RIF1 on a centromeric plasmid allowed cdc13-1 rif1Δ cells to grow at 25°C ( Figure 1B ) . The synthetic effect of the cdc13-1 rif1Δ combination was not uncovered during a previous genome wide search for gene deletions enhancing the temperature-sensitivity of cdc13-1 cells [37] , likely because that screening was done at 20°C , a temperature at which cdc13-1 rif1Δ double mutants do not show severe growth defects ( Figure 1A ) . Our data above indicate that Rif1 , but not Rif2 , is required to support cell viability when Cdc13 protective function is partially compromised . If the lack of Rif1 in cdc13-1 cells increased the temperature-sensitivity by exacerbating the telomere end protection defects of these cells , Rif1 overexpression might suppress the temperature sensitivity caused by the cdc13-1 allele . Indeed , high copy number plasmids carrying wild type RIF1 , which had no effect on wild type cell viability , improved the ability of cdc13-1 cells to form colonies on synthetic selective medium at the semi-permissive temperature of 26–27°C ( Figure 1C ) . The function of Cdc13 in telomere protection is mediated by its direct interactions with Stn1 and Ten1 , leading to formation of the CST complex ( reviewed in [38] ) . In addition to the capping function , the CST complex is implicated in repression of telomerase action [2] , [21] , [23] , [24] . This CST-dependent negative regulation of telomerase can be separated from CST capping function , as yeast cells either carrying the cdc13-5 allele or lacking the Stn1 C-terminus ( residues 282–494 ) ( stn1ΔC ) display extensive telomere elongation but no or minimal growth defects [24]–[26] . We evaluated the specificity of the genetic interaction between rif1Δ and cdc13-1 by analysing the consequences of deleting RIF1 and RIF2 in cdc13-5 or stn1ΔC cells . Deletion of RIF1 turned out to reduce cell viability of cdc13-5 mutant cells at any temperatures , while deletion of RIF2 did not ( Figure 2A ) . Furthermore , meiotic tetrad dissection of stn1ΔC/STN1 rif1Δ/RIF1 diploid cells did not allow the recovery of viable stn1ΔC rif1Δ double mutant spores ( Figure 2B ) , indicating that rif1Δ and stn1ΔC were synthetic lethal . By contrast , viable rif2Δ stn1ΔC spores were found with the expected frequency after tetrad dissection of stn1ΔC/STN1 rif2Δ/RIF2 diploid cells ( Figure 2C ) . The observed synthetic phenotypes suggest that both stn1ΔC and cdc13-5 cells have capping deficiencies and that the lack of Rif1 enhances their protection defects . Consistent with this hypothesis , cdc13-5 and stn1ΔC mutants were shown to accumulate telomeric ssDNA , although the amount of this ssDNA was not enough to invoke a DNA damage response [24] , [25] . We conclude that Rif1 , but not Rif2 , is required to support cell viability when a partial inactivation of CST capping function occurs . A Cdc13 specific function that is not shared by the other subunits of the CST complex is its requirement for recruitment/activation of telomerase at chromosome ends [2]–[6] . Cdc13-mediated telomerase recruitment is disrupted by the cdc13-2 mutation , which leads to progressive telomere shortening and senescence phenotype [4] . We therefore asked whether RIF1 deletion influences viability and/or senescence progression of cdc13-2 cells . Viable cdc13-2 rif1Δ spores were recovered after tetrad dissection of cdc13-2/CDC13 rif1Δ/RIF1 diploid cells ( data not shown ) , indicating that the lack of Rif1 does not affect the overall viability of cdc13-2 cells . When spores from the dissection plate were streaked on YEPD plates for 4 successive times , the decline in growth of cdc13-2 and cdc13-2 rif1Δ spores occurred with similar kinetics ( Figure 2D ) , indicating that RIF1 deletion did not accelerate the senescence phenotype of cdc13-2 cells specifically defective in telomerase recruitment . Taken together , these genetic interactions indicate that Rif1 , but not Rif2 , has a role in assisting the essential function of the CST complex in telomere protection . The CST complex functionally and physically interacts with the polα-primase complex [7] , [21] , [22] , [25] , which is essential for telomeric C-strand synthesis during telomere elongation . Thus , we analyzed the genetic interactions between rif1Δ and temperature sensitive alleles affecting DNA primase ( pri2-1 ) [39] or polα ( cdc17-1 and pol1-1 ) [40] , [41] . Both cdc17-1 rif1Δ and pol1-1 rif1Δ cells were viable , but their temperature-sensitivity was greatly enhanced compared to cdc17-1 and pol1-1 single mutants ( Figure 2E ) . Similarly , the maximal permissive temperature of the pri2-1 rif1Δ double mutant was reduced relative to that of pri2-1 single mutant cells ( Figure 2F ) . Moreover both pol1-1 rif1Δ and pri2-1 rif1Δ cells showed growth defects even at the permissive temperature of 25°C ( Figure 2E and 2F ) . Thus , Rif1 , like CST , functionally interacts with the polα-primase complex . The synthetic effects of combining rif1Δ with cdc13 and stn1 mutations suggest that Rif1 might normally assist the Cdc13 and Stn1 proteins in carrying out their essential telomere protection functions . It is known that cdc13-1 cells undergo checkpoint-dependent metaphase arrest when incubated at the restrictive temperature [14] . Failure to turn on the checkpoint allows cdc13-1 cells to form colonies at 28°C [42] , [43] , indicating that checkpoint activation can partially account for the loss of viability of cdc13-1 cells . We then asked whether the enhanced temperature sensitivity of cdc13-1 rif1Δ cells compared to cdc13-1 cells might be due to upregulation of the DNA damage checkpoint response . Deletion of the checkpoint gene RAD9 , which partially suppressed the temperature sensitivity of cdc13-1 mutant cells , slightly improved the ability of cdc13-1 rif1Δ cells to grow at 23–25°C ( Figure 3A ) , indicating that the synthetic interaction between Rif1 and Cdc13 can be partially alleviated by checkpoint inactivation . Furthermore , when wild type , rif1Δ , cdc13-1 and cdc13-1 rif1Δ cell cultures were arrested in G1 with α-factor at 20°C ( permissive temperature ) and then released from G1 arrest at 25°C ( non-permissive temperature for cdc13-1 rif1Δ cells ) , they all replicated DNA and budded with similar kinetics after release ( Figure 3B and 3C ) . However , most cdc13-1 rif1Δ cells then arrested in metaphase as large budded cells with a single nucleus , while wild type , cdc13-1 and rif1Δ cells divided nuclei after 75–90 minutes ( Figure 3D ) . To assess whether the cell cycle arrest of cdc13-1 rif1Δ cells was due to DNA damage checkpoint activation , we examined the Rad53 checkpoint kinase , whose phosphorylation is necessary for checkpoint activation and can be detected as changes in Rad53 electrophoretic mobility . Rad53 was phosphorylated in cdc13-1 rif1Δ cells that were released from G1 arrest at 25°C , whereas no Rad53 phosphorylation was seen in any of the other similarly treated cell cultures ( Figure 3E ) . RIF1 deletion caused a checkpoint-mediated G2/M cell cycle arrest also in cdc13-5 cells . In fact , exponentially growing cdc13-5 rif1Δ cell cultures at 25°C contained a higher percentage of large budded cells with a single nucleus than rif1Δ or cdc13-5 cell cultures under the same conditions ( Figure 3F ) . Furthermore , Rad53 phosphorylation was detected in these cdc13-5 rif1Δ cells , but not in the rif1Δ and cdc13-5 cell cultures ( Figure 3G ) . Thus , the lack of Rif1 results in DNA damage checkpoint activation in both cdc13-1 and cdc13-5 cells under conditions that do not activate the checkpoint when Rif1 is present . The lack of Rif1 is known to cause telomere overelongation [29] . Thus , we examined telomere length in cdc13-1 rif1Δ double mutant cells . The length of duplex telomeric DNA was examined after transferring at 25°C cell cultures exponentially growing at 20°C , followed by Southern blot analysis with a TG-rich probe of XhoI-digested genomic DNA prepared at different times after shift at 25°C ( Figure 4A ) . As expected [29] , rif1Δ mutant cells had longer telomeres than wild type and cdc13-1 cells ( Figure 4A ) . Telomeres in cdc13-1 rif1Δ double mutant cells either at 20°C or after incubation at 25°C were longer than those of wild type and cdc13-1 cells , but undistinguishable from those of rif1Δ cells ( Figure 4A ) . Not only RIF1 deletion , but also the cdc13-5 mutation is known to cause telomere overelongation [24] ( Figure 4B ) . Interestingly , when telomere length was analyzed in cdc13-5 rif1Δ double mutant cells grown at 25°C , telomeres were longer in cdc13-5 rif1Δ double mutant cells than in cdc13-5 and rif1Δ single mutants ( Figure 4B ) , indicating that the cdc13-5 mutation exacerbates the telomere overelongation defect caused by the lack of Rif1 . The finding that telomeres in cdc13-1 rif1Δ double mutant cells at 25°C were longer than those of cdc13-1 cells , but undistinguishable from those of cdc13-1 rif1Δ cells grown at 20°C ( Figure 4A ) suggests that the growth defects of cdc13-1 rif1Δ cells at 25°C are not due to rif1Δ-induced telomere overelongation . Telomere lengthening in rif1Δ mutant cells is telomerase-dependent [44] and requires the action of the checkpoint kinase Tel1 that facilitates telomerase recruitment [45] , [46] . To provide additional evidences that loss of viability in cdc13 rif1Δ mutants occurs independently of rif1Δ-induced alterations in telomere length , we asked whether RIF1 deletion was still deleterious in cdc13-1 , cdc13-5 and stn1ΔC cells in a context where telomeres cannot be elongated due to the lack of Tel1 [45] . We found that TEL1 deletion did not alleviate the growth defects of cdc13-1 rif1Δ cells ( Figure 5A ) . Rather , cdc13-1 tel1Δ and cdc13-1 rif1Δ tel1Δ cells showed an enhanced temperature sensitivity compared to cdc13-1 and cdc13-1 rif1Δ cells , respectively , presumably due to the combined effects of loss of a telomere elongation mechanism and inability to protect telomeres from shortening activities . Furthermore , the growth defects of cdc13-5 rif1Δ double mutant cells were similar to those of cdc13-5 rif1Δ tel1Δ triple mutant cells ( Figure 5B ) . Finally , viable stn1ΔC rif1Δ tel1Δ mutant spores could not be recovered after meiotic tetrad dissection of stn1ΔC/STN1 rif1Δ/RIF1 tel1Δ/TEL1 diploid cells ( data not shown ) , indicating that stn1ΔC and rif1Δ were synthetic lethal even in the absence of Tel1 . As telomere lengthening is dramatically increased when both Rif1 and Rif2 are absent [30] , we also investigated whether the absence of Rif2 exacerbates cdc13-1 rif1Δ growth defects . As shown in Figure 5C , cdc13-1 rif1Δ rif2Δ cells formed colonies at the maximum temperature of 20°C and behaved similarly to cdc13-1 rif1Δ cells . We therefore conclude that the synthetic interaction between rif1Δ and cdc13 alleles is not due to rif1Δ-induced alterations in telomere length , but it is a direct consequence of Rif1 loss . It is known that cdc13-1 cells at 37°C accumulate telomeric ssDNA that triggers checkpoint-mediated cell cycle arrest [14] . Thus , we investigated whether cdc13-1 rif1Δ and cdc13-5 rif1Δ cells contained aberrant levels of single-stranded TG sequences at their telomeres that could be responsible for loss of viability in cdc13-1 rif1Δ and cdc13-5 rif1Δ cells at 25°C . The integrity of chromosome ends was analyzed by an in-gel hybridization procedure [9] , probing for the presence of single-stranded TG sequences . Both cdc13-1 and rif1Δ single mutants either grown at 20°C ( Figure 6A , lanes 2 and 4 ) or incubated at 25°C for 3 hours ( Figure 6A , lanes 6 and 8 ) showed only a very slight increase in single-stranded TG sequences compared to wild type ( Figure 6A , lanes 1 and 5 ) . By contrast , cdc13-1 rif1Δ double mutant cells contained higher amounts of telomeric ssDNA than cdc13-1 and rif1Δ cells already at 20°C ( Figure 6A , lane 3 ) and the amount of this ssDNA increased dramatically when cdc13-1 rif1Δ cells were incubated at 25°C for 3 hours ( Figure 6A , lane 7 ) . A similar telomere deprotection defect was observed also for cdc13-5 rif1Δ cells grown at 25°C ( Figure 6A , lane 11 ) , which displayed an increased amount of telomeric ssDNA compared to similarly treated wild type and cdc13-5 cells ( Figure 6A , lanes 9 and 10 ) . Because the length of single-stranded G overhangs increases during S phase [8] , the strong telomeric ssDNA signals observed in cdc13-1 rif1Δ cell cultures at 25°C ( Figure 6A ) might be due to an enrichment of S/G2 cells . We ruled out this possibility by monitoring the levels of single-stranded TG sequences in cdc13-1 rif1Δ cell cultures that were arrested in G2 with nocodazole at 20°C and then transferred to 25°C in the presence of nocodazole for 3 hours ( Figure 6B ) . Similarly to what we observed in exponentially growing cell cultures , G2-arrested cdc13-1 rif1Δ cells at 20°C displayed increased amounts of ssDNA compared to each single mutant under the same conditions , and incubation at 25°C led to further increase of this ssDNA ( Figure 6B ) . Taken together , these findings indicate that the lack of Rif1 causes a severe defect in telomere protection when Cdc13 activity is partially compromised . If telomeric ssDNA accumulation contributes to checkpoint activation in cdc13-1 rif1Δ and cdc13-5 rif1Δ cells , then mutations reducing ssDNA generation should alleviate the arrest and relieve the lethality caused by the lack of Rif1 in cdc13-1 and cdc13-5 background . Because the Exo1 nuclease contributes to generate telomeric ssDNA in cdc13-1 cells [47] , we examined the effect of deleting EXO1 in cdc13 rif1Δ cells . When G2-arrested cell cultures at 20°C were transferred to 25°C for 3 hours , cdc13-1 rif1Δ exo1Δ triple mutant cells contained significantly lower amounts of telomeric ssDNA than cdc13-1 rif1Δ cells ( Figure 6B ) . A similar behaviour of the triple mutant was detectable even when G2-arrested cultures where kept at 20°C , although the quantity of telomeric ssDNA accumulated by cdc13-1 rif1Δ cells at this temperature was lower than at 25°C ( Figure 6B ) . Furthermore , EXO1 deletion partially suppressed both the temperature-sensitivity of cdc13-1 rif1Δ cells ( Figure 7A ) and the loss of viability of cdc13-5 rif1Δ cells ( Figure 7B ) , further supporting the hypothesis that reduced viability in these strains was due to defects in telomere protection . Exo1-mediated suppression of the cdc13 rif1Δ growth defects correlated with alleviation of checkpoint-mediated cell cycle arrest . In fact , when cell cultures exponentially growing at 20°C were incubated at 25°C for 3 hours , the amount of both metaphase-arrested cells and Rad53 phosphorylation was reproducibly lower in cdc13-1 rif1Δ exo1Δ cells than in cdc13-1 rif1Δ cells ( Figure 7C and 7D ) . Similar results were obtained also with cdc13-5 rif1Δ exo1Δ cells growing at 25°C , which accumulated less metaphase-arrested cells and phosphorylated Rad53 than similarly treated cdc13-5 rif1Δ cells ( Figure 7E and 7F ) . Thus , both cell lethality and checkpoint-mediated cell cycle arrest in cdc13 rif1Δ cells appear to be caused , at least partially , by Exo1-dependent telomere DNA degradation . The lack of Rif1 might increase the lethality of cells with reduced CST activity just because it causes a telomere deprotection defect that exacerbates the inherent telomere capping defects of cdc13 or stn1 mutants . If this hypothesis were correct , RIF1 deletion should affect viability also of other non-CST mutants defective in end protection . Alternatively , Rif1-CST functional interaction might be specific , thus reflecting a functional connection between Rif1 and CST . To distinguish between these two possibilities , we analyzed the effects of deleting RIF1 in Yku70 lacking cells , which display Exo1-dependent accumulation of telomeric ssDNA , as well as checkpoint-mediated cell cycle arrest at elevated temperatures ( 37°C ) [47] , [51] . Loss of Yku in est2Δ cells , which lack the telomerase catalytic subunit , leads to synthetic lethality , presumably due to the combined effects of telomere shortening and capping defects [48]–[50] , [52] . As expected [53] , yku70Δ cells were viable at 25°C and 30°C , but they were unable to form colonies at 37°C ( Figure 8A ) . Similarly , yku70Δ rif1Δ double mutant cells grew well at 25°C and 30°C ( Figure 8A ) and did not show Rad53 phosphorylation when grown at 25°C ( Figure 8B , time 0 ) . Furthermore , similar amounts of phosphorylated Rad53 were detected in both yku70Δ and yku70Δ rif1Δ cell cultures that were kept at 37°C for 4 hours ( Figure 8B ) , indicating that loss of Rif1 does not enhance the telomere protection defects already present in yku70Δ cells . Consistent with a previous observation [54] , RIF1 deletion partially suppressed the temperature sensitivity ( Figure 8A ) and the telomere length defect ( data not shown ) caused by the lack of Yku70 , suggesting that the elongated state of the telomeres could be the reason why yku70Δ rif1Δ cells can proliferate at 37°C . Checkpoint activation can also be induced during telomere erosion caused by insufficient telomerase activity [55] , [56] . Thus , we asked whether RIF1 deletion accelerated senescence progression and/or upregulated checkpoint activation in cells lacking the telomerase catalytic subunit Est2 . Meiotic tetrads were dissected from a diploid strain heterozygous for the est2Δ and rif1Δ alleles , which are recessive and therefore do not affect telomere length in the diploid . After 2 days of incubation at 25°C ( approximately 25 generations ) , spore clones from the dissection plate were both streaked for 4 successive times ( Figure 8C ) and propagated in YEPD liquid medium to prepare protein extracts for Rad53 phosphorylation analysis at different time points ( Figure 8D ) . Similar to what was previously observed [44] , RIF1 deletion did not accelerate senescence progression in est2Δ cells , as est2Δ rif1Δ clones showed a decline in growth similar to that of est2Δ clones ( Figure 8C ) . Furthermore , est2Δ and est2Δ rif1Δ cell cultures showed similar patterns of Rad53 phosphorylation with increasing number of generations ( Figure 8D ) . Thus , the lack of Rif1 does not enhance either DNA damage checkpoint activation or senescence progression during telomere erosion caused by the lack of telomerase . Finally , because the telomerase machinery is known to be recruited to an unrepaired DSB [57] , we ruled out the possibility of a general role for Rif1 in inhibiting checkpoint activation by examining activation/deactivation of the checkpoint induced by an unrepaired DSB . To this end , we used JKM139 derivative strains , where a single DSB can be generated at the MAT locus by expressing the site-specific HO endonuclease gene from a galactose-inducible promoter [58] . This DSB cannot be repaired by homologous recombination , because the homologous donor sequences HML or HMR are deleted . As shown in Figure 8E , when G1-arrested cell cultures were spotted on galactose containing plates , both wild type and rif1Δ JKM139 derivative cells overrode the checkpoint-mediated cell cycle arrest within 24–32 hours , producing microcolonies with 4 or more cells . Moreover , when galactose was added to exponentially growing cell cultures of the same strains , Rad53 phosphorylation became detectable as electrophoretic mobility shift in both wild type and rif1Δ cell cultures about 2 hours after HO induction , and it decreased in both cell cultures after 12–15 hours ( Figure 8F ) , when most cells resumed cell cycle progression ( data not shown ) . Thus , Rif1 does not affect the checkpoint response to an irreparable DSB . Altogether these data indicate that Rif1 supports specifically CST functions in telomere protection .
Both shelterin and CST complexes are present in a wide range of unicellular and multicellular organisms , where they protect the integrity of chromosomes ends ( reviewed in [38] ) . Thus , the understanding of their structural and functional connections is an important issue in telomere regulation . We have approached this topic by analysing the consequences of disabling the shelterin-like S . cerevisiae proteins Rif1 or Rif2 in different hypomorphic mutants defective in CST components . We provide evidence that Rif1 , but not Rif2 , is essential for cell viability when the CST complex is partially compromised . In fact , RIF1 deletion exacerbates the temperature sensitivity of cdc13-1 mutant cells that are primarily defective in Cdc13 telomere capping functions . Furthermore , cells carrying the cdc13-5 or the stn1ΔC mutation , neither of which causes per se DNA damage checkpoint activation and growth defects [24] , [26] , grow very poorly or are unable to form colonies , respectively , when combined with the rif1Δ allele . By contrast , RIF1 deletion does not affect either viability or senescence progression of cdc13-2 cells , which are specifically defective in telomerase recruitment . This Cdc13 function is not shared by the other CST subunits , suggesting that Rif1 is specifically required to support the essential capping functions of the CST complex . Cell lethality caused by the absence of Rif1 in both cdc13-1 and cdc13-5 cells appears to be due to severe telomere integrity defects . In fact , telomeres in both cdc13-1 rif1Δ and cdc13-5 rif1Δ double mutant cells display an excess of ssDNA that leads to DNA damage checkpoint activation . Deleting the nuclease EXO1 gene partially restores viability of cdc13-1 rif1Δ and cdc13-5 rif1Δ cells and reduces the level of telomeric ssDNA in cdc13-1 rif1Δ cells , indicating that cell lethality in cdc13 rif1Δ cells is partially due to Exo1-dependent telomere DNA degradation and subsequent activation of the DNA damage checkpoint . Although Rif1 and Rif2 interact both with the C-terminus of Rap1 and with each other [29] , [30] , our finding that only Rif1 is required for cell viability when Cdc13 or Stn1 capping activities are reduced indicates that Rif1 has a unique role in supporting CST capping function that is not shared by Rif2 . Earlier studies are consistent with the idea that Rif1 and Rif2 regulate telomere metabolism by different mechanisms [30] , [31] , [35] . Furthermore , while the content of Rif2 is lower at shortened than at wild type telomeres , the level of Rif1 is similar at both , suggesting that these two proteins are distributed differently along a telomere [59] . Finally , inhibition of telomeric fusions requires Rif2 , but not Rif1 [32] . Noteworthy , although RIF1 deletion is known to cause telomere overelongation [29] , the synthetic interaction between Rif1 and CST occurs independently of rif1Δ-induced alterations in telomere length . In fact , the lack of Tel1 , which counteracts rif1Δ-induced telomere overelongation [45] , does not alleviate the growth defects of cdc13 rif1Δ cells . Furthermore , deletion of RIF2 , which enhances telomere elongation induced by the lack of Rif1 [30] , does not exacerbate the synthetic phenotypes of cdc13 rif1Δ double mutant cells . Thus , loss of viability in cdc13 rif1Δ cells is not due to telomere overelongation caused by RIF1 deletion , but it is a direct consequence of Rif1 loss . By analyzing the effects of combining RIF1 deletion with mutations that cause telomere deprotection without affecting CST functions , we found that the functional interaction between Rif1 and the CST complex is highly specific . In fact , the lack of Rif1 does not enhance the DNA damage checkpoint response in telomerase lacking cells , which are known to experience gradual telomere erosion leading to activation of the DNA damage checkpoint [55] , [56] . Furthermore , RIF1 deletion does not upregulate DNA damage checkpoint activation in yku70Δ cells , which display Exo1-dependent accumulation of ssDNA and checkpoint-mediated cell cycle arrest at 37°C [47]–[51] . This is consistent with previous observations that comparable signals for G strand overhangs can be detected on telomeres derived from yku70Δ and yku70Δ rif1Δ cells [54] , indicating that RIF1 deletion does not exacerbate the end protection defect due to the absence of Yku . By contrast , the lack of Rif1 partially suppresses both temperature-sensitivity and telomere shortening in yku70Δ cells ( Figure 8A ) [54] , possibly because the restored telomere length helps to compensate for yku70Δ capping defects . Notably , although RIF1 deletion leads to telomere overelongation in cdc13-1 and cdc13-5 mutants , this elongated telomere state does not help to increase viability in cdc13-1 rif1Δ and cdc13-5 rif1Δ cells . The simplest interpretation of the specific genetic interactions we found between Rif1 and CST is that a functional connection exists between Rif1 and the CST complex , such that Rif1 plays a previously unanticipated role in assisting the CST complex in carrying out its essential telomere protection function . Indeed , this functional interaction is unexpected in light of Rif1 and CST localization along a telomere . In fact , while CST is present at the very ends of chromosomes , Rif1 is thought to be distributed centromere proximal on the duplex telomeric DNA [59] . However , as yeast telomeres have been proposed to fold back onto the subtelomeric regions to form a ∼3-kb region of core heterochromatin [60] , [61] , this higher-order structure could place Rif1 and CST in close proximity , thus explaining their functional interaction . The function of Rif1 in sustaining CST activity cannot be simply attributable to the Rif1-mediated suppression of ssDNA formation at telomeres , as rif1Δ cells show only a very slight increase in ssDNA at both native ( Figure 6 ) and HO-induced telomeres [33] compared to wild type . Furthermore , although deletion of Rif2 leads to increased amounts of telomeric ssDNA [33] , cdc13-1 rif2Δ , cdc13-5 rif2Δ and stn1ΔC rif2Δ double mutants are viable and do not display growth defects . Finally , other mutants defective in telomere capping or telomere elongation ( yku70Δ and est2Δ ) are perfectly viable in the absence of Rif1 . One possibility is that Rif1 physically interacts , directly or indirectly , with the CST complex . Indeed , human Stn1 was found to copurify with the shelterin subunit TPP1 [62] , suggesting the existence of CST-shelterin complexes in mammals . Unfortunately , we were so far unable to coimmunoprecipitate Rif1 with Cdc13 or Stn1 , and further analyses will be required to determine whether Rif1 and the CST complex undergo stable or transient association during the cell cycle . Indeed , not only 5′-3′ resection , but also incomplete synthesis of Okazaki fragments is expected to increase the size of the G tail during telomere replication . The yeast CST complex genetically and physically interacts with the polα-primase complex [7] , [22] , [25] and the human CST-like complex increases polα-primase processivity [63] , [64] . Furthermore , the lack of CST function in G1 and throughout most of S phase does not lead to an increase of telomeric ssDNA [13] , suggesting that the essential function of CST is restricted to telomere replication in late S phase . Altogether , these observations suggest that CST may control overhang length not only by blocking the access of nucleases , but also by activating polα-primase-dependent C-strand synthesis that can compensate G tail lengthening activities . Based on the finding that Rif1 regulates telomerase action and functionally interacts with the polα-primase complex ( Figure 2 ) , it is tempting to propose that Rif1 favours CST ability to replenish the exposed ssDNA at telomeres through activation/recruitment of polα-primase , thus coupling telomerase-dependent elongation to the conventional DNA replication process . The recent discoveries that human TPP1 interacts physically with Stn1 [62] and that CST-like complexes exist also in S . pombe , plants and mammals [65]–[68] raise the question of whether functional connections between the two capping complexes exist also in other organisms . As telomere protection is critical for preserving genetic stability and counteracting cancer development , to address this question will be an important future challenge .
Strain genotypes are listed in supplementary Table S1 . Unless otherwise stated , the yeast strains used during this study were derivatives of W303 ( ho MATa ade2-1 his3-11 , 15 leu2-3 , 112 trp1-1 ura3 can1-100 ) . All gene disruptions were carried out by PCR-based methods . The cdc13-1 mutant was kindly provided by D . Lydall ( University of Newcastle , UK ) . The cdc13-2 mutant was kindly provided by V . Lundblad ( Salk Institute , La Jolla , USA ) . The stn1ΔC and cdc13-5 alleles carried a stop codon following amino acids 282 and 694 respectively [24] , [25] , and were generated by PCR-based methods . Wild type and cdc13-1 strains carrying either the 2 µ vector or 2 µ RIF1 plasmid were constructed by transforming wild type and cdc13-1 strains with plasmids YEplac195 ( 2 µ URA3 ) and pML435 ( 2 µ RIF1 URA3 ) , respectively . The strains used for monitoring checkpoint activation in response to an irreparable DSB were derivatives of strain JKM139 ( MATa ho hmlΔ hmrΔ ade1 lys5 leu2-3 , 112 trp1::hisG ura3-52 ade3::GAL-HO ) , kindly provided by J . Haber ( Brandeis University , Waltham , MA , USA ) [58] . To induce HO expression in JKM139 and its derivative strains , cells were grown in raffinose-containing yeast extract peptone ( YEP ) and then transferred to raffinose- and galactose-containing YEP . Cells were grown in YEP medium ( 1% yeast extract , 2% bactopeptone , 50 mg/l adenine ) supplemented with 2% glucose ( YEPD ) or 2% raffinose ( YEP+raf ) or 2% raffinose and 2% galactose ( YEP+raf+gal ) . Synthetic complete medium lacking uracil supplemented with 2% glucose was used to maintain the selective pressure for the 2 µ URA3 plasmids . Genomic DNA was digested with XhoI . The resulting DNA fragments were separated by electrophoresis on 0 . 8% agarose gel and transferred to a GeneScreen nylon membrane ( New England Nuclear , Boston ) , followed by hybridization with a 32P-labelled poly ( GT ) probe and exposure to X-ray sensitive films . Standard hybridization conditions were used . Visualization of single-stranded overhangs at native telomeres was done by in-gel hybridization [9] , using a single-stranded 22-mer CA oligonuleotide probe . The same DNA samples were separated on a 0 . 8% agarose gel , denatured and hybridized with an end-labeled C-rich oligonucleotide for loading control . For western blot analysis , protein extracts were prepared by TCA precipitation . Rad53 was detected using anti-Rad53 polyclonal antibodies kindly provided by J . Diffley ( Clare Hall , London , UK ) . Secondary antibodies were purchased from Amersham and proteins were visualized by an enhanced chemiluminescence system according to the manufacturer . Flow cytometric DNA analysis was determined on a Becton-Dickinson FACScan on cells stained with propidium iodide . | Protection of chromosome ends is crucial for maintaining chromosome stability and genome integrity , and its failure leads to genome rearrangements that may facilitate carcinogenesis . This protection is achieved by the packaging of chromosome ends into protective structures called telomeres that prevent DNA repair/recombination activities . Telomeric DNA is bound and stabilized by two protein complexes named CST and shelterin , which are present in a wide range of multicellular organisms . Whether structural and functional connections exist between these two capping complexes is an important issue in telomere biology . Here , we investigate this topic by analyzing the consequences of disabling the two Saccharomyces cerevisiae shelterin-like components , Rif1 and Rif2 , in different hypomorphic mutants defective in CST components . We demonstrate that Rif1 plays a previously unanticipated role in assisting the essential telomere protection function of the CST complex , indicating a tight coupling between CST and Rif1 . As CST complexes have been recently identified also in other organisms , including humans , which all rely on shelterin for telomere protection , this functional link between CST and shelterin might be an evolutionarily conserved common feature to ensure telomere integrity . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"dna",
"metabolism",
"microbiology",
"telomeres",
"model",
"organisms",
"dna",
"chromosome",
"biology",
"biology",
"molecular",
"biology",
"cell",
"biology",
"nucleic",
"acids",
"yeast",
"and",
"fungal",
"models",
"saccharomyces",
"cerevisiae",
"dna",
"recombination",
"molecular",
"cell",
"biology"
] | 2011 | Rif1 Supports the Function of the CST Complex in Yeast Telomere Capping |
The absence of an effective vaccine and the debilitating chemotherapy for Leishmaniasis demonstrate the need for developing alternative treatments . Several studies conducted with Morinda citrifolia have shown various biological activities , including antileishmanial activity , however its mechanisms of action are unknown . This study aimed to analyze the in vivo activity of M . citrifolia fruit juice ( Noni ) against Leishmania ( Leishmania ) amazonensis in C57BL/6 mice . M . citrifolia fruit juice from the Brazilian Amazon has shown the same constitution of other juices produced around the world and liquid chromatography–mass spectrometry analysis identified five compounds: deacetylasperulosidic acid , asperulosidic acid , rutin , nonioside B and nonioside C . Daily intragastric treatment with Noni was carried out after 55 days of L . ( L . ) amazonensis infection in C57BL/6 mice . Parasitic loads , cytokine and extracellular protein matrix expressions of the lesion site were analyzed by qPCR . Histopathology of the lesion site , lymph nodes and liver were performed to evaluate the inflammatory processes . Cytokines and biochemical parameters of toxicity from sera were also evaluated . The Noni treatment at 500 mg . kg-1 . day-1 for 60 days decreased the lesion size and parasitic load in the footpad infected with L . ( L . ) amazonensis . The site of infection also showed decreased inflammatory infiltrates and decreased cytokine expressions for IL-12 , TNF-α , TGF-β and IL-10 . On the other hand , Noni treatment enhanced the extracellular matrix protein expressions of collagen IV , fibronectin and laminin in the infected footpad as well collagen I and II , fibronectin and laminin in the mock-infected footpads . No toxicity was observed at the end of treatment . These data show the efficacy of Noni treatment .
Leishmaniasis is one of the seventeen neglected diseases prioritized by the World Health Organization . Although most cases of neglected diseases are in underdeveloped countries , leishmaniasis is spreading worldwide [1] . The infection caused by Leishmania parasites may remain asymptomatic or evolve to a symptomatic form that can vary from a cutaneous to a visceral form of the disease , the latter of which can be lethal if left untreated [2] . As there is no vaccine against leishmaniasis yet , infected people are treated with antileishmaniasis drugs and control still depends on programs focusing on the vector and reservoir hosts [1 , 3] . There are a limited number of drugs for the treatment of leishmaniasis and the pentavalent antimonials are the most common [3] . However , antimonials can cause severe adverse effects , such as vomiting , nausea , anorexia , myalgia , abdominal pain , headache , arthralgia , and lethargy , due to their accumulation in the tissues [1] . Until now , efforts to reduce the toxicity of antileishmaniasis drugs have been unsuccessful , which reinforces the need for new antileishmanial drugs . Therefore , protocols that could provide an alternative therapy , reduce dosages , treatment duration and adverse effects for leishmaniasis , would be welcome . Morinda citrifolia Linn . is a small plant native to Southeast Asia . It is commonly known as Noni and is one of the most significant resources of traditional medicine in S . E . Asian countries . The efficacy of Noni in the treatment of pain and inflammatory reactions [4] as well as its antimicrobial activity [5] has been demonstrated in various studies . Recently , morindicone and morinthone , isolated from the stem of M . citrifolia , were shown to have activity , in vitro , against Leishmania ( L . ) major [6] . In order to demonstrate the antileishmanial activity of M . citrifolia , our group has been using the fruit juice of this plant in in vitro assays with Leishmania ( L . ) infantum promastigotes and intracellular amastigotes . Our previous results showed cytoplasmic vacuolization , lipid inclusion , increased exocytosis activity and autophagosome-like vesicles in L . ( L . ) infantum promastigotes treated with M . citrifolia fruit juice . Cytotoxicity assay with J774 . G8 macrophages showed that M . citrifolia fruit juice was not toxic to these cells up to 1000μg . mL-1; however , when intracellular amastigotes were evaluated by light microscopy , macrophages showed vacuoles with probable remains of intracellular parasites [7] . Based on these results , the aim of the present study was to evaluate the antileishmanial activity of M . citrifolia fruit juice under in vivo conditions , using C57BL/6 mice subcutaneously infected with L . ( L . ) amazonensis .
Morinda citrifolia fruits were collected in São Luiz ( S2°31 W44°16 ) , a municipality in the Brazilian Amazon , located 24m above sea level . Fully ripe fruits , with a translucent exocarp , were picked in the rainy season , from April to November 2011 . The material was properly identified by Ana Maria Maciel Leite and the voucher specimen number 2000346 was deposited at the Herbarium Professora Rosa Mochel at the Universidade Estadual do Maranhão . Fruits were washed with sterilized distilled water , dried at 25°C and placed in sterile glass bottles for 3 days to drain off the extract . The juice extract , called Noni , from M . citrifolia fruit was centrifuged twice at 4000 rpm for 15 minutes; the supernatant was lyophilized and stored at -20°C . Noni was dissolved in PBS immediately before use in the in vivo experiments . Lyophilized noni was dissolved in methanol to 5mg . mL-1 . The LC Shimadzu Nexera UFLC was coupled to an ion trap Bruker Amazon . Analyses were performed at ambient temperature in a 100mm x 2 . 1mm x 2 . 6μm Kinetex C18 gravity column , equipped with an 8 mm x 4 mm , 5μm guard column . The mobile phase consisted of water containing 0 . 1% formic acid ( eluent A ) and acetonitrile ( eluent B ) . The gradient of B was as follows: in 5 . 5 min from 5% to 25% , from 7 . 0 to 8 . 5 min up to 100% B , held at 100% for 1 . 5 min , then 100% to 5% in 1 min , and finally held at 5% for 2 min . The flow rate was 0 . 3 mL/min and the injection volume was 1 μL . Other specifications were as described in the literature [8] . Female C57BL/6 mice 4-6-weeks old were obtained from Centro de Criação de Animais de Laboratório ( CECAL/FIOCRUZ ) and maintained under pathogen-free conditions , controlled temperature and food and water ad libitum . All experiments with animals were conducted in accordance with the guidelines for experimental procedures of the Conselho Nacional de Controle de Experimentação Animal ( CONCEA ) and approved by Comissão de Ética no Uso de Animais from Fundação Oswaldo Cruz ( CEUA-FIOCRUZ ) , identification number LW72/12 . The L . ( L . ) amazonensis ( MHOM/BR/1976/MA-76 ) obtained from a human case of diffuse infection and characterized by isoenzyme [9] and lectin techniques [10] was maintained in the laboratory by successive passages in BALB/c mice . Prior to infection , parasites were isolated from a non-ulcerated nodular lesion in the footpad and amastigote viability was checked with erythrosine B by light microscopy . 104 amastigote forms were inoculated subcutaneously into the right footpad of C57BL/6 mice . Initially , an 8-week pilot treatment protocol , with two different concentrations of Noni ( 250 and 500mg . kg-1 ) , was carried out to determine the dose of Noni to be used in the posterior analyses . The daily treatment was carried out with 100μL of Noni by gavage . A group of non-treated infected mice was maintained as control . Lesion thickness was evaluated weekly in order to choose the most efficient drug concentration . Treatment protocol was performed with 5 groups of 10 animals , as follows: infected and treated ( 100μL of Noni 500mg . kg-1 by gavage , daily ) ; infected and control drug-treated ( Glucantime 20mg . kg-1 by intramuscular injection , twice a week ) ; infected and mock-treated ( 100μL of PBS by gavage , daily ) ; mock-infected and treated ( 100μL of Noni 500mg . kg-1 by gavage , daily ) ; and normal ( mock-infected and mock-treated ) . Treatment started 55 days after infection for all groups . Lesion kinetics was evaluated weekly by a caliper rule , in comparison to the non-infected contralateral footpad and expressed as lesion thickness . After 30 and 60 days of treatment animals were euthanized , blood was collected to obtain serum and tissue fragments from footpad , draining lymph nodes and liver were excised for posterior analyses . DNA from the footpad and draining lymph nodes of 3 animals per group was extracted following a standard phenol/chloroform protocol [11] . DNA concentration was quantified in a NanoDrop 2000c spectrophotometer ( ThermoScientific ) . Parasite load was estimated by real time PCR performed in Applied Biosystems Step One Plus equipment , using Fast SYBR Green Master Mix . Primers were target for the parasite kDNA and mouse β-actin was used as an endogenous control ( S1 Table ) . Skin , lymph nodes and liver fragments were fixed in 10% buffered formalin and routinely processed for paraffin embedding . Tissue sections ( 5μm thick ) were stained with Hematoxylin-Eosin , Gomori trichrome and Picrosirius red . Tissues were observed under a light microscope and polarized light was used to observe the collagen fibers . After euthanasia , skin fragments of infected footpads from 3 mice of each group were collected . Total RNA was extracted using TRIZOL reagent ( Invitrogen , Karlsruhe , Germany ) following the manufacturer’s instructions . cDNA synthesis was performed with 1μg of total RNA using a iScript cDNA Synthesis kit ( Bio-Rad Laboratories , Hercules , CA ) according to the manufacturer’s recommendations . Primers targeting the genes IL-4 , IL-10 , IL-12 , TNF-α , IFN-γ , TGF-β , iNOS , Laminin , Fibronectin and Collagens I , III and IV were designed using the Primer Express software version 3 . 0 ( Applied Biosystems , 2004 ) , and manufactured by Invitrogen ( Supplementary Data 1 ) . Real Time PCR assays were performed using Power SYBR Green Master Mix and the relative quantification ( 2-ΔΔCT ) method was applied , using the mouse RPLP0 gene ( large ribosomal protein , P0 ) as the endogenous control . Results were analyzed with the StepOne Software v2 . 3 ( Applied Biosystems ) . A pool of sera obtained from the blood of five mice per group was used for cytokine quantification of IL-4 , IL-10 , IL-12 , TNF-α , IFNγ ( BD Bioscience ) and TGF-β ( R&D System ) following the manufacturer's specifications . Clinical signs of toxicity , such as piloerection , diarrhea , salivation , convulsions or changes in mobility , respiration rate or muscle tone , were observed during the treatments . Levels of alanine transaminase ( ALT ) , aspartate transaminase ( AST ) , alkaline phosphatase ( ALP ) , total protein , direct bilirubin , indirect bilirubin , total bilirubin , albumin , globulin , urea and creatinine were analyzed in sera pools from mice treated for 60 days in Ciba Corning equipment . At necropsies , stomach and gut mucosa were macroscopically evaluated for abnormal findings . Animal weight was measured on an analytical balance after 30 and 60 days of treatment . The values were expressed as mean ± S . D . The results were analyzed statistically by Analysis of Variance ( ANOVA ) followed by Bonferroni’s post-test . The analyses were performed with the software GraphPad Prism 5 . 0 . 4 . Differences were considered significant when p<0 . 05 .
According to the selective ions and elution order obtained from the Liquid Chromatography–Mass spectrometry analysis and compared with references in the literature [8] , five compounds were identified: deacetylasperulosidic acid ( 1 ) , asperulosidic acid ( 2 ) , rutin ( 3 ) , nonioside B ( 4 ) and nonioside C ( 5 ) ( Fig 1 ) . The extract ion chromatograms ( m/z ) of these compounds were respectively 389 , 431 , 609 , 629 and 467 . The pilot protocol showed that Noni at 500mg . kg-1 could significantly reduce lesion growth from the fourth week of treatment . Therefore , the dosage of 500mg . kg-1 was chosen for subsequent protocols . In this protocol , the treatment was able to significantly reduce lesion size as of the sixth week , when compared with the infected non-treated group ( Fig 2A ) . The control drug , Glucantime 20mg . kg-1 , was also able to decrease lesion size , showing no statistical difference with Noni treatment . After 30 days of Noni treatment there was no change but after 60 days the parasite loads in the footpad and draining lymph node had significantly decreased in comparison to the non-treated control , corroborating with the results of the lesion kinetics ( Fig 2B and 2C ) . On the other hand , Glucantime was able to reduce parasite loads after 30 days of treatment for the footpad and after 60 days for the lymph node . Histopathological analysis of the lesion site of mock-treated mice showed inflammatory infiltrates composed of parasitized macrophages 30 and 60 days after infection ( Fig 3 ) . In the former , the infiltrated area had increased and the number of infected macrophages enhanced . Furthermore , a large area of necrosis and lesion ulcerations was observed . Also at 30 days after infection the draining lymph node presented hyperplasia of the cortical region . Noni treatment at 30 and 60 days reduced the parasite loads and inflammatory infiltrate . Remarkable tissue remodeling at the lesion site and depletion of the number of blast cells in the lymph node were observed after 60 days of treatment . Also , at that time , no parasites were found in the Glucantime-treated mice at the lesion site . A reduction of the inflammatory infiltrate was also noted in the skin as well as the reestablishment of the normal histopathological pattern of the lymph node . When the expressions of IFN-γ , iNOS , IL-12 , TNF-α , IL-10 , TGF-β and IL-4 at the lesion site were evaluated , no difference was noted between Noni and mock-treated non-infected mice ( Fig 4 ) . In mock treated infected mice , IFN-γ and iNOS showed upregulation after 30 days of treatment but this was not observed on the 60th day . In Noni treated infected mice this upregulation was not observed and mice showed the normal value throughout the experiment . However , Glucantime treated infected mice presented an upregulation of these cytokines principally on the 30th day of treatment . An upregulation of IL-4 was also noted at this time in Glucantime and mock treated infected mice . The IL-12 , TNF-α and IL-10 expressions were upregulated in mock treated infected mice especially after the 60th day . This upregulation was not observed in infected and Noni or Glucantime treated mice . TGF-β was upregulated throughout the experiment in mock treated infected mice . However , TGF-β was upregulated only after 60 days of treatment in Noni or Glucantime treated mice . Cytokine levels in the serum showed that L . ( L . ) amazonensis increased IL-4 and TNF-α at 30 days and IL-10 at both treatment times ( Fig 5 ) . This high production of TNF-α was also observed in the infected groups treated with Noni or Glucantime . IL-10 showed a lower increase in infected and treated groups . On the other hand , IL-4 production decreased in treated groups whether infected or not . No alterations were observed in IL-12 or TGF-β except for a slight increase in TGF-β production for the Noni treated infected mice . Finally , Glucantime treatment increased IFN production after 60 days . Observation of the lesion site skin stained with hematoxylin-eosin showed a reduction in the normal structure of the dermis and a degradation of the connective tissue in infected footpads when compared with mock-infected groups . This difference in the presence of collagen fibers among the groups was demonstrated using Gomori´s trichrome and Picrosirius Red ( Fig 6A ) . To quantify these alterations , the extracellular matrix protein expression was evaluated by qPCR in footpad skin after 60 days of treatment ( Fig 6B–6F ) . Noni fruit juice upregulated the expression of all analyzed proteins , except for collagen IV in mock-infected mice; whereas L . ( L . ) amazonensis downregulated the expression of fibronectin , collagen I and IV when compared with normal mice . Noni and Glucantime treatment preserved the normal expression of collagen IV , laminin and fibronectin in infected footpads but they decreased collagen I and III expression . No clinical signs of toxicity were observed during the treatments and there was no mortality . There was no significant statistical alteration in the weight of the animals after 30 and 60 days of treatment . During necropsy , alterations such as hyperemia were not observed in the stomach or gut mucosa of the animals treated with Noni . Also there was no change in the sera biochemical parameters of hepatic and renal functions , except for alanine transaminase ( ALT ) . Infection and Noni treatment enhanced ALT levels but within the normal maximum limit . Histopathology showed that Noni treatment in mock-infected mice did not stimulate an inflammatory reaction in the liver ( Fig 7 ) . The L . ( L . ) amazonensis on the contrary , induced a diffuse and periportal inflammatory infiltration , the latter being reduced by Noni or Glucantime treatments .
M . citrifolia has various biological actions including leishmanicidal [6] and immunomodulatory activities [12 , 13] that have not yet been fully elucidated . The chromatographic analysis of the Noni juice used in our studies showed the same pattern of other Noni juices produced around the world [8] . It is translucent and brown; presents medium viscosity , characteristic odor , pH 3 . 9 and yielded 6 . 31% of a highly hygroscopic powder [7] . In this study we used Noni juice to treat C57BL/6 mice infected with L . ( L . ) amazonensis . We chose to begin the treatment 55 days after infection , when the lesion was well established , in order to better mimic treatment in humans . In fact , when treatment began , all lesions were about 2mm thick . Noni treatment decreased the lesion size associated with a lower parasite load in the skin and draining lymph nodes after 60 days of treatment . Treatment with the control drug , Glucantime , caused a faster reduction of the parasite load than Noni . However , after 60 days of treatment , Noni had reduced the lesion size more than Glucantime . The lesion size reduction after Noni treatment is associated with a decreased parasite load and control of the inflammatory process caused by L . ( L . ) amazonensis . The histopathology and cytokine expression analysis showed a reduction in focal inflammation in the skin after Noni treatment with a downregulation of cytokine expressions ( IL-12 and TNF-α ) at 30 and 60 days of treatment . IFN-γ plays a crucial function in controlling the Leishmania infection , as has been demonstrated in mice with genetic defects in this molecule and/or its receptor [14] . IFN-γ induces parasite elimination by activating both phagocyte oxidase ( phox ) and iNOS , which is the most effective mechanism of killing intracellular parasites mediated by macrophages [15 , 16] . In vitro , our group demonstrated an increase of nitric oxide production and iNOS expression in the peritoneal macrophages infected with L . ( L . ) amazonensis and treated with Noni [17] . In the present study , the association of high levels of IFN-γ and iNOS and decrease of parasite load was observed in Glucantime treatment , but not in Noni treated mice , suggesting a different mechanism of parasite killing in vivo . Our results demonstrated that normal levels of IL-10 expression in treated groups were associated to low parasite burden , while high levels of IL-10 expression were associated to elevated parasite burden in mock-treated infected mice showing the role of IL-10 in maintaining the infection . The same results have been described in IL-10 knock-out mice which were more competent in controlling L . ( L . ) major infection than cells from wild type mice [18] . The TGF-β expression was also upregulated in infected mice . Proteins secreted by infected macrophages or the promastigote forms of L . ( L . ) infantum chagasi activate the soluble form of latent TGF-β complex favoring the persistence of parasites within infected macrophages through induction of TGF-β mediated anti-inflammatory mechanisms [19] . The role of TGF-β as a key predictive factor of enhanced susceptibility to the disease was also demonstrated in BALB/c mice immunized with whole antigens of L . ( L . ) amazonensis . Species-specific components of vaccine activate TGF-β production that predisposes more susceptible individuals to a more aggravated form of the disease [20] . Thus , a low TGF-β expression in Noni treated and infected mice contributes to maintain the control of inflammatory infiltrates when compared with infected mock-treated mice . The phenotype of susceptibility in L . ( L . ) major infection is clearly associated to high levels of IL-4 and Th2 response [21] . IL-4 reduces iNOS expression and enhances disease progression due to increased survival and growth of Leishmania parasites in infected cells [22] . In our work , L . ( L . ) amazonensis enhanced the IL-4 expression as expected , while treatment with Noni maintained lower levels of IL-4 expression in infected mice . As the treatment was performed by gavage , the amount of cytokines in sera allows us to verify the immunomodulatory effect of Noni . In addition , L . ( L . ) amazonensis infection is not limited to the skin . The parasite tends to disseminate to the lymph nodes and can even reach the spleen and liver [23] . The cytokines measured in sera revealed an enhancement of IL-4 and IL-10 caused by L . ( L . ) amazonensis infection , which were not seen after Noni or Glucantime treatment . The decreased levels of IL-4 and IL-10 contribute to maintain a Th1 response in the treated groups . Furthermore , the increase of IFN-γ levels at 60 days due to Glucantime treatment contributes to the effectiveness of the macrophages by iNOS induction in skin and parasite load decrease . Indeed , Noni treatment decreased the IL-4 levels even in mock-infected mice; this may be due to the activation of cannabinoid 2 receptors [12] . Altogether , these results endorse the immunomodulatory effects of Noni . Studies have reported that an immunochemotherapy is more effective than chemotherapy or immunotherapy [24] , and our data show that Noni treatment is actually immunochemotherapy . In addition to cytokine modulation , the skin histopathology analysis showed that Noni helps to control the inflammatory infiltrates and supports an early remodeling process . The tissue repair process is critically important for rapid cure of cutaneous leishmaniasis , as demonstrated in L . ( L . ) major murine cutaneous leishmaniasis [25] , and is associated to reduced IL-10 and increased TNF-α , IFN-γ [26] and the TGF-β pathway [27] . In a non-infected wound , high levels of IL-10 decrease pro-inflammatory mediators and inflammation , normal collagen deposition and restoration of normal dermal architecture [28] , whilst TGF-β induces immune cell recruitment , promotes matrix protein synthesis while decreasing matrix protein degradation leading to fibrotic tissue formation [29] . In contrast , Noni treatment promotes a control of the inflammatory process which contributes to a favorable ambient for tissue repair . The increase in TGF-β levels after 60 days of treatment with Noni , when compared to 30 days of treatment , may be associated with this tissue repair . The excessive secretion of pro-inflammatory cytokines and chemokines , as observed in mock-treated infected mice , can recruit and activate additional inflammatory cells and lead to uncontrolled tissue degradation , including new granulation tissue and growth factors , delaying collagen deposition , which impairs the repair process and perpetuates the non-healing condition [30] . The histopathological evaluation of collagen fibers and protein expression in the skin confirmed a modulation of extracellular matrix proteins in Noni-treated mock-infected mice . Anthraquinones were previously identified in our Noni juice [7] and an anthraquinone isolated from Noni fruit has been shown to stimulate collagen type I , the major component of extracellular matrix of the skin in human dermal fibroblasts . Nano-emulsion with this anthraquinone increased the dermal procollagen I in nude mouse skin [31] in the same way as the Noni increased collagen I expression in mock-infected mice . Moreover , the overexpression of collagen III , laminin and fibronectin by Noni treatment is reported here for the first time . The extracellular matrix protein expressions most affected by Leishmania infection were collagen I , collagen IV and fibronectin . The role of collagen I , IV and fibronectin during Leishmania infection have been well described in the literature . L . ( L . ) mexicana binds fibronectin and collagen I to promote adhesion and phagocytosis by macrophages [32 , 33] . Degradation of fibronectin and collagen IV by glycoprotein gp63 seems to enhance L . ( L . ) amazonensis migration . Leishmania-degraded fibronectin by surface and secreted leishmanolysin also decreases the production of reactive oxygen intermediates by parasite-infected macrophages and affects the accumulation of intracellular parasites [34 , 35] . Treatment with Noni or Glucantime restored the collagen IV and fibronectin expressions to normal levels . This is possibly due to the reduction of parasitic burden and control of the inflammation process with Noni treatment . In addition , Noni treatment also caused an upregulation of laminin expression , a protein related to the degradation and binding of Leishmania [36] . Finally , the toxicity parameters analyzed in our model indicated that Noni treatment has no toxic effect on mice . No alterations in the mucosa of stomach or gut were found , showing that the Noni juice does not irritate the digestive system . This result was expected since a previous work described that M . citrifolia had a preventive effect on gastro-esophageal inflammatory diseases [37] . Although there was a slight increase in ALT , which did not exceed the normal limits , there was a decrease in the hepatic inflammation caused by L . ( L . ) amazonensis . Nevertheless , Noni toxicity still needs more studies , considering the controversial data in literature that sometimes show toxicity [38 , 39] , no toxicity [40–42] or even a liver protective effect [43] . The present work has proved the efficacy of Noni juice in reducing the parasite burden and lesion size . In addition , it has shown its modulatory effects on cytokine and extracellular matrix protein expressions . Altogether , Noni treatment has an antileishmanial activity , associated with an immunomodulatory action , which opens a new path to follow in the quest to promote a rapid clinical cure of cutaneous leishmaniasis . | Leishmaniasis is a complex of diseases caused by parasites of the Leishmania genus , which affects thousands of people around the world . The parasite lives within the cells and the disease manifests itself in different ways , one of them is wound-like lesions on the skin that do not heal . The treatment , with a medicament discovered in 1912 , causes several side effects , its intramuscular administration is painful and it is given daily over a long period of time . These characteristics show the need for new alternatives for leishmaniasis treatment justifying works like this . The Morinda citrifolia is a plant native to the Polynesian islands and has a fruit commonly known as Noni . Noni has been analyzed for various targets such as anticancer , anti-inflammatory and antimicrobial effects; however , the antileishmanial has not yet been fully evaluated . This work proves that Noni treatment can promote a fast clinical cure in mice with leishmaniasis by decreasing parasite number , acting on the immune system and repairing skin components . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [
"blood",
"cells",
"innate",
"immune",
"system",
"medicine",
"and",
"health",
"sciences",
"immune",
"physiology",
"cytokines",
"pathology",
"and",
"laboratory",
"medicine",
"immune",
"cells",
"immunology",
"parasitic",
"diseases",
"parasitic",
"protozoans",
"collagens",
"developmental",
"biology",
"lymph",
"nodes",
"protozoans",
"signs",
"and",
"symptoms",
"leishmania",
"lymphatic",
"system",
"molecular",
"development",
"white",
"blood",
"cells",
"inflammation",
"animal",
"cells",
"proteins",
"immune",
"response",
"immune",
"system",
"biochemistry",
"diagnostic",
"medicine",
"anatomy",
"cell",
"biology",
"physiology",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"macrophages",
"organisms",
"extracellular",
"matrix",
"proteins"
] | 2016 | Morinda citrifolia Linn. Reduces Parasite Load and Modulates Cytokines and Extracellular Matrix Proteins in C57BL/6 Mice Infected with Leishmania (Leishmania) amazonensis |
Local field potential ( LFP ) oscillations are often accompanied by synchronization of activity within a widespread cerebral area . Thus , the LFP and neuronal coherence appear to be the result of a common mechanism that underlies neuronal assembly formation . We used the olfactory bulb as a model to investigate: ( 1 ) the extent to which unitary dynamics and LFP oscillations can be correlated and ( 2 ) the precision with which a model of the hypothesized underlying mechanisms can accurately explain the experimental data . For this purpose , we analyzed simultaneous recordings of mitral cell ( MC ) activity and LFPs in anesthetized and freely breathing rats in response to odorant stimulation . Spike trains were found to be phase-locked to the gamma oscillation at specific firing rates and to form odor-specific temporal patterns . The use of a conductance-based MC model driven by an approximately balanced excitatory-inhibitory input conductance and a relatively small inhibitory conductance that oscillated at the gamma frequency allowed us to provide one explanation of the experimental data via a mode-locking mechanism . This work sheds light on the way network and intrinsic MC properties participate in the locking of MCs to the gamma oscillation in a realistic physiological context and may result in a particular time-locked assembly . Finally , we discuss how a self-synchronization process with such entrainment properties can explain , under experimental conditions: ( 1 ) why the gamma bursts emerge transiently with a maximal amplitude position relative to the stimulus time course; ( 2 ) why the oscillations are prominent at a specific gamma frequency; and ( 3 ) why the oscillation amplitude depends on specific stimulus properties . We also discuss information processing and functional consequences derived from this mechanism .
Recent experiments under awake conditions indicate that there is little effective spatial contrast in the firing rate of neurons in olfactory structures [1] and even in auditory primary sensory structures [2] , which suggests a limited role for the mean firing rate in sensory function . In addition , it has recently been shown in the retina that the sole firing rate is not sufficient to code for behavioral performance [3] . Furthermore , the role of spike timing for plasticity [4] and coding [5] suggests that the temporal structure of neuronal activity may be crucial for perception . In particular , various functional studies have reported that fast local field potential ( LFP ) oscillations , particularly those in the gamma band ( 40–80 Hz ) that correlate with perception [6] and attention [7] , simultaneously induce a greater synchrony in the firing of cells , thereby having a greater impact on downstream structures [7] . In the mammalian olfactory bulb ( OB ) , odorant stimulation induces LFP oscillations both in the gamma ( 40–80 Hz ) and beta ( 15–35 Hz ) ranges . There is , however , significant disparity in the reports regarding the conditions in which these oscillations are expressed . In anesthetized animals , gamma and beta LFP oscillations are odor-induced and appear alternately along the respiratory cycle , with gamma bursts specifically occurring during the inspiration/expiration ( I/E ) transition [8] , [9] . In the awake rat , gamma and beta LFP oscillations exist spontaneously , and odors evoke increases [10] or decreases [11] of amplitude in the gamma frequency range . In insects [5] , [12] and to a lesser extent in fish [13] , spiking activity has been shown to be strongly linked to the oscillation which plays therefore a crucial role in coding . In rodents the role of these oscillation has been primarily described to reflect the experience of the animal [14] , [15] . Despite the findings of phase/time relationships between mitral ( MC ) spiking activity and the oscillation or between MC pairs [16]–[18] , a better description of these relationships across the MC population is needed before demonstrating that such temporal activity can support an odor code . Our approach focused on gamma oscillations in anesthetized rats , which are not very different from oscillations recorded under behaving conditions [19] . The aim of the current study was two-fold: ( 1 ) to analyze the fine temporal relationships between the spiking of an MC population and gamma oscillations in anesthetized , freely-breathing rats in response to odorant stimulation and ( 2 ) to build a biophysical model of MC activity under the same in vivo conditions to explain the data . Our data showed that during the transient periods of gamma oscillation , a specific subset of MCs exhibited specific periodic firing phase patterns that could be grouped in a small number of types . Furthermore , the pattern types exhibited by a particular MC are qualitatively related to the nature of the odorant stimulus . These different types of patterns constitute the signature of an entrainment of these cells by an input oscillation [20] , [21] , whose origin can be attributed to the inhibitory granular input [17] . In the model , we verify that this scenario of entrainment effectively reproduces the experimental observations . We further explored the model's dependence on MC input activity and intrinsic properties , showing that our model robustly exhibited an optimum of entrainment when the oscillation frequency was within the gamma range . Overall , these results provide a detailed quantitative description of the complex MC firing activity at the population level and , at the same time , a theoretical explanation of the links between MC spike trains and gamma LFP oscillations evoked under freely breathing conditions in the OB . Furthermore , these results constitute an important underlying mechanism if future research shows that such a firing activity can support odor coding in behaving conditions .
The relationships between unit activity and gamma LFP oscillations during phase-locking raise the question of whether reciprocal entrainment mechanisms exist between MC and LFP oscillations . We addressed one part of this question using a biophysical model 1 ) to investigate the modalities of entrainment of MC activities by granule cell activities and 2 ) to explain the spike train structures measured experimentally . The use of a model allowed us to perform studies that are not currently feasible with animal studies .
Our work provides converging arguments in favor of a MC entrainment , suggesting that they receive an oscillatory input in vivo . Although several groups have focused on the temporal relationship between spikes and LFP , the exploration of this entrainment phenomenon has only been partly addressed . The analysis of Eeckman and Freeman [16] differed due to the multi-unit nature of their signal . Kashiwadani et al . [18] focused more on a coding hypothesis rather than a dynamic hypothesis and used an artificial sniff , which likely affected the bulbar dynamics . One in vitro study has explored the spike/LFP relationship [17] . The network connectivity and dynamic conditions , however , could be very different from those in vivo , and it is difficult to compare in vitro oscillations to those under in vivo conditions . One potential criticism of our work is the usefulness of such a mechanism under some behavioral conditions , such as where odor presentation induces a decrease in gamma oscillations [14] . The fact that the amplitude of gamma oscillations decreases under particular conditions does not indicate that gamma oscillations are unnecessary for the system . Indeed , the amplitude decreases under particular conditions; however , beta oscillations are simultaneously enhanced , which suggests that the olfactory bulb and cortex together work in a different manner [27] , [28] . There are other behavioral conditions in which gamma oscillations do not decrease but rather increase with odor presentation [10] . Experimental conditions vary widely depending on the employed paradigm , which can induce different sniffing strategies [29] or different neuromodulatory control of the MC/GC loop [30] . These different conditions are certainly responsible for the variation in the expression of gamma and beta oscillations . The aim of our study was to explore how spikes and oscillation can be phase-locked in the more general situations in which gamma oscillations are present . It would be an interesting second step to perform the same study for beta oscillations and to look for another mechanism that could be responsible for the observed spike/beta oscillation synchronization [31] . Our MC model was fitted to reproduce the quantitative and dynamic properties of MCs , based upon recent findings on MC membrane activity . It is consistent with experimental data [17] , [32]–[34] and includes essential features of existing models [35]–[37] . In our model , entrainment was produced by a relatively weak oscillatory inhibitory conductance . When the cell has an input well above the threshold , it fires regularly and even periodically in response to a constant input . Under these conditions , and when the oscillatory input is sufficiently weak , the neuronal response can be predicted from the phase response curve ( PRC ) [38] , [39] , which can also predict some characteristics of the tongues [40] . Since the PRC of our MC model agrees with the experimental one [41] , it is likely that our model can accurately predict the MC response to such synaptic inputs . The synaptic mechanisms hypothesized in our model are based on observations made by Lagier et al . ( 2004 ) [17] , who reported that a fraction of the IPSPs received by the MCs were phase-locked relative to gamma oscillations . Other possible mechanisms that could affect the MC membrane potential during gamma oscillations have been proposed: ( 1 ) a synchronous inhibition resulting from GABA release from astrocytes [42] , ( 2 ) a synchronous excitation of the MCs by glutamate spillover onto lateral dendrites [43] , [44] , and ( 3 ) a direct influence of the LFP on membrane voltage [45] , [46] . Of these possible explanations , the functional and dynamic relevance of the interactions between MCs and GCs [47] are the most likely processes to occur during gamma oscillations . Aside from this observation , an important characteristic of our model is that the time constant of the single inhibitory inputs does not play a role because it is the frequency of the oscillatory modulation of the whole inhibitory input that imposes the locking frequency . It is important to insert this mechanism into a more realistic model in order to take into consideration the spatial distribution of the physiological mechanism in dendrites . The presence of high frequency bursts of spikes in MCs is probably not due solely to the excitatory drive from the sensory input , but also to additional conductance levels beyond those present in the model . For example , the activation of an intrinsic conductance , such as T channels , would facilitate the emergence of high frequency bursts like those in our recordings . Neuronal phase-locking and pattern formation have already been described both theoretically [20] , [21] , [48] and experimentally [26] , [49] . Our study , performed in the context of the OB with a resonant neuron , however , sheds light on important aspects that are specific to this system . First , the simulation results provide some precise explanations regarding electrophysiological properties that are not directly accessible by experimentation . The phase of the MC spikes relative to LFP oscillations is precisely distributed in the ¼ to ½ part of the LFP cycle due to their level of synaptic drive . The phase distributions extracted from simulated data ( Fig . 6A ) are more precise and go beyond the data that was obtained in the OB [14] , [17] , [18] , as well as our own experimental results . We wish to point out , however , that our experimental measurement of gamma phase was not totally reliable from one particular depth to another , and could have up to ¼ of a cycle of phase shift . Indeed , gamma oscillations are generated by membrane currents flowing on either side of the MC layer , and the simultaneous recording of a gamma oscillation constantly shows a ¼ cycle phase shift when it is recorded up to 100 µm from the MC layer ( Fourcaud-Trocmé and Buonviso , personal communication ) . In our recordings , we could not control the exact position of each electrode relative to the MC layer . As a consequence , these recording constraints , as well as the noise level , are likely to be the source of the larger spread of phases during the phase-locked experimental pattern when compared to the model prediction ( see Fig . 6B ) . This can also explain the small discrepancies with recordings from other studies in which the LFP recording position relative to the mitral cell layer was different or more precisely controlled [16]–[18] . In all cases , however , our results are still qualitatively comparable . Second , lower levels of excitatory conductance in the model were predicted to induce a slight phase lag ( corresponding to ∼5 ms ) between the neurons , as shown in Fig . 6A . This phase difference is observed in the cross-correlograms that were presented by Kashiwadani et al . ( 1999 ) [18] ( their Fig . 5A ) , and it may be achieved by differential activation of the glomerulus [18] , [50] , [51] . Our predictions are able to explain the result of this previous study . The model predicts that phase-locked or residual spike trains may depend on the degree of detuning ( i . e . , the difference between the neuronal intrinsic firing rate and the external network frequency ) , the strength of coupling , and the level of noise . The continuity between the locked and residual spike trains suggests that entrainment of MCs may be fine-tuned during the gamma oscillation , and may corroborate the locking that was observed in the fish OB [13] . However , only simple phase-locking has been found in these species . Finally , in our model , increasing the global level of excitatory-inhibitory conductance favors entrainment ( see Fig . 7A–B ) . Under anesthetized conditions , the maximal conductance should be directly related to the maximal activity of the peripheral excitation [51] , [52] and to the maximum of synaptic inhibition [53] , which corresponds to the inspiration/expiration transition . Taken together , these data imply that entrainment should be the strongest at this transition , which is what was observed with the gamma wave appearing around this point ( Fig . 1 ) . Interestingly , odors that elicit low theta activity ( respiratory modulation ) , which potentially reflects weak OB activation ( e . g . , low-concentration odors ) , generally failed to elicit gamma oscillation [9] , suggesting that the theta oscillation could gate gamma activity . In the OB , MCs are directly interconnected only within glomerulus [54]–[57] and not across bulbar areas . Instead , MC spikes induce a depolarization of some GC spines , which propagates through parts of the cell to other spines and triggers both synchronous [58]–[62] and asynchronous [36] , [55] , [63]–[65] GABA release from granule dendrodendritic synapses onto other MCs . This peak of GABA release is about ¾ of the gamma cycle ( ∼ = 12 ms ) after the peak of spiking ( according to the spike ( Fig . 4E ) and IPSP ( Fig . 4G ) phase distributions in [17] ) . The slow component of inhibition [55] , [64]–[66] may contribute to the global tonic inhibition level that is present in the OB . This synaptic activity suggests that the effective interaction between MCs is inhibition , and that the OB could be approximately modeled as a network of inhibitory coupled MCs . Our results raise the question of the origin of gamma oscillatory activity in the OB . Do entrained MCs participate in the creation of this rhythm ? If so , how ? Two models have been recently proposed [35] , [36] . In [35] , a network of inhibitory coupled MCs is applied to in vitro recordings , in which the gamma oscillation is continuous and evoked by single stimulation . In [36] , the particular tendency of MCs to synchronize their firing when their stochastic inputs become correlated was studied . Neither of these models , however , has been shown to induce the spiking activity that we report here . This is also the case for the general mechanisms that have been proposed for oscillation generation in inhibitory [67]–[69] and excitatory-inhibitory networks [69]–[71] . It has to be noted that , in these studies , a large heterogeneity in the population intrinsic firing rates usually prevents the formation of an oscillation , and that the case of a resonant neuron is not considered . Our results , however , suggest the following two hypotheses . First , if MCs behave like oscillators , a possible mechanism for the emergence of the oscillation may be provided by the transition towards synchrony , a phenomenon that is generally observed in a network of coupled heterogeneous oscillators [20] , [72] . Second , the model predicts that intrinsic MC properties ( according to Fig . 7D and 8 ) lead to a maximal entrainment in the gamma frequency range . Indeed , a non resonant neuron ( i . e . an integrate-and-fire model ) does not exhibit this behavior ( data not shown ) . From this property , it can be seen that among the MC population that exhibits quite heterogeneous frequencies ( see Fig . 4 ) , MCs firing with intrinsic firing rate in the range of 40–70 Hz are better entrained than MCs firing at lower or higher frequencies . The rhythm could be created when this entrained MC population becomes sufficiently large . Other studies in anesthetized rats have shown that MCs can be segregated into two populations according to whether they phase-lock to gamma oscillations or to beta oscillations , which is likely due to their position relative to the receptive field of the odor [31] . Here , we show that , among MCs that lock to the gamma oscillation , locking is finely tuned depending on the various conditions that control MC entrainment , one of them being the nature of the odor . A subset of the MC population can therefore be entrained and phase-locked at gamma frequencies with patterns depending on the intrinsic firing rate of the MC , which is similar to the previously described rate-specific synchrony [26] . The resulting phase-locked activity map may overlap with the glomerular activity map . This is suggested by our results , which show that pattern types are partially reproducible for the same cell and the same odor . The glomerular map is likely forwarded to MCs as a “rate code” map that is readable by downstream structures . The phase-locked activity map , however , appears to be more elaborate than the simple glomerular map . In particular , the main characteristic of the phased-locked activity map is that it can be modulated by various parameters: such as the conductance level ( Fig . 7 ) , the MC intrinsic firing rate ( Figs . 4 and 5 ) , and/or the degree of GC-MC coupling ( Fig . 5C ) . It is therefore likely that this phase-locked activity map is modifiable by plasticity mechanisms and central control [10] . Based on the adjustment of the spike phases , the mechanisms described here would result in population activity that is readable by decoding structures which function as coincidence detectors . Pyramidal cells of the piriform cortex have such detection properties [73] . In addition , synaptic integration by pyramidal dendrites is more sensitive to pattern-like inputs , as shown in other sensory systems [74] , [75] . Thus , our model of a phase-locked map of MC activity across the OB supports the idea developed by Zou and Buck [76] that pyramidal cells could integrate odorant features using a combination of coincident MC inputs ( as argued also in [77] ) in a way similar to the mushroom bodies of insects [78] , [79] .
All experiments were performed in accordance with the guidelines of the European Communities Council . Male Wistar rats ( 150–350 g ) obtained from Charles River Labs ( L'Arbresle ) were anesthetized with urethane ( i . p . 1 . 5 mg/kg , with additional supplements as needed ) and placed in a stereotaxic apparatus . The dorsal region of the OB was exposed . Bulbar activity was recorded as a broadband signal ( 0 . 1–5 kHz ) using 16-channel silicon probes ( NeuroNexus Technologies , Ann Arbor , MI ) with a homemade , 16-channel DC amplifier . The data were digitally sampled at 10 kHz and acquired on a PC using the IOtech acquisition system ( Wavebook , IOtech , Cleveland , OH ) . Probes were placed in the lateral or medial part of the OB at a depth that maximized the number of channels located in the mitral cell layer ( MCL ) . The respiration signal was recorded using a homemade flowmeter based on a fast response time thermodilution airflow sensor . Odors were delivered through a dilution olfactometer ( 440 ml/min ) . The recording protocol was as follows: 5 s of spontaneous activity , 5 s of odor-evoked activity , and 5 s of post-stimulus activity . Each sampling included stimulation by simple linear aliphatic compounds . The varying features of odorants were either the number of carbons in the main chain ( chain length: 5 , 6 , 7 and 10 C ) or the functional group associated with the chain ( alcohol , ester , aldehyde or ketone ) . Additional odors have been used , including isoamyl acetate , p-cymene , and eugenol . All odors were delivered in front of the animal's nose at a fraction of 18 . 10−2 of the saturated vapor pressure . The time delay between each odor presentation was at least one minute . The MC model was derived from a previous study [35] , which was derived in turn from [85] . The model uses parameters from [86] , [87] . Its essential features are ( 1 ) spiking activity through sodium spikes , ( 2 ) bursting activity ( Fig . 11A ) , ( 3 ) its current-frequency response ( Fig . 11B ) , ( 4 ) resonant properties as revealed through sub-threshold oscillations ( Fig . 11C ) , and ( 5 ) phase response curve ( Fig . 11D ) . The method to compute the resonance frequency is to linearize the system of dynamical equations around its equilibrium for a given membrane potential and to extract this frequency from the imaginary part of complex eigenvalue λ ( Imag ( λ ) /2π ) . The method to compute the phase response curve [39] was to impose a small instantaneous voltage deviation along the period of spiking ( here equal to 25 ms ) . We reduced the number of variables in the model from nine to four by removing each of the variables individually and checking that the simpler model retained the essential features listed above; at the same time , some of the parameters were adjusted as necessary . The main characteristics were preserved as depicted in Fig . 11 . The remaining variables were the membrane potential ( V ) , the activation gating variable of the fast rectifying potassium current ( mKf ) , the activation ( mKs ) and the inactivation ( hKs ) gating variable of the slow potassium current . The dynamics is described in Eqs . 5–8: ( 5 ) ( 6 ) ( 7 ) ( 8 ) Isyn is the synaptic current and is detailed later . The gating variable parameters are given by: The maximal conductance of the ionic channels was gNa = 500 S/m2; gNaP = 1 . 1 S/m2; gKs = 310e S/m2; gKA = 100 S/m2; gKf = 100 S/m2 . The leak conductance is gL = 0 . 1 S/m2 . The product is approximated to 0 . 004 . The reverse ionic potentials are EK = −70 mV and ENa = 45 mV , and the membrane reverse potential is EL = −66 . 5 mV . The membrane capacitance is 0 . 01 F/cm2 . In terms of amplitude , to assess the synaptic inputs that the MCs receive under in vivo conditions , the MC model input resistance was verified to be consistent with the order of magnitude of synaptic inputs that an MC actually receives . Experimentally , the input resistance of an MC was found to be 115 MΩ [88] . We assumed that a limited part of the total membrane area was responsible for the input resistance and that axial resistance ( or intracellular resistivity ) was negligible for this neuronal small portion . This portion included the soma and proximal dendrites and corresponded to 10% of the total area or ∼15 . 103 µm2 [89] , [90] given that , beyond a distance of about 100 µm from the soma on either the primary or secondary dendrite , synaptic inputs are not conveyed to the soma [91] . This experimental “effective” input resistance becomes 1 . 72 Ω . m2 when related to the membrane unit area . Our model is consistent with experiments since its input resistance measured at hyperpolarized membrane potential is 1 . 5 Ω . m2 . Previous studies have examined the kinetics of signaling to MCs and have found that MCs receive slow [64] , fast , and asynchronous or simply fast inhibition from GCs [60] . In [17] , it was reported that MCs receive weakly phased inhibition during gamma oscillation . From Fig . 4G of their report , we visually estimated that the amplitude of this phasic inhibition was up to 30% . The preferential phase of this oscillatory inhibition was estimated to be 359±9° of the LFP oscillation , and it showed strong statistical significance ( p<10−8 by the Rayleigh test ) . Therefore , based on these experimental observations , our model assumes that on average , MCs receive an input current of the form ( Eq . 9 ) : ( 9 ) which is composed of a baseline conductance , including constant excitatory ( gE ) and inhibitory ( gI ) conductances , as well as a sinusoidal part , i . e . , the oscillating conductance whose amplitude is gIo and whose frequency is fosc . A white noise component was added to each excitatory and inhibitory component with amplitudes gEs and gIs , respectively , that were chosen to induce noise comparable to what is observed with in vivo intracellular voltage recording [88] . No other source of noise was added . gEs and gIs vary proportionally with gE and gI in Figs . 5 and 6 . gEs is set to zero in Figs . 7 and 8 . gIs is set to 0 . 282 S . m−2 . ms1/2 in Fig . 7B and 0 . 141 S . m−2 . ms1/2 in Fig . 7D . The reversal potentials of the excitatory and inhibitory synapses are EE = 0 mV and EI = −70 mV , corresponding to the reversal potentials of AMPA/NMDA receptors and GABAA receptors , respectively . Inhibitory synaptic inputs can completely stop MC firing activity [50] and reshape the temporal organization of spike trains [60] . It was therefore appropriate for our model to take into account the levels of inhibitory conductance that could have such effects on MC activity . The maximum conductance of a single synaptic input has been estimated to be on the order of 1 nS [92] , [93] . A single MC has around 10 , 000 connections with GCs . As mentioned above , only about 10% of these inputs are conveyed to the soma [91] , which corresponds to ∼1 , 000 synapses . Such a limited integration of the total synaptic input is consistent with other studies [94] , [95] . Assuming that an MC receives inhibitory inputs from 1 , 000 synapses during every gamma cycle ( period of ∼16 ms ) and assuming an exponential time decay of 5 ms for GABAa [53] , [60] , it can be shown that , during the gamma oscillation , the MC average inhibitory input conductance is given by 1000*1nS*5 ms/16 ms = 312 nS . For the membrane area mentioned above , this represents gI = 20 S/m2 . gE was adapted to reach a frequency up to 200 Hz , as observed in vivo . For example , Fig . 11E shows a voltage trace of the MC model that is forced by the oscillation and that receives tonic excitatory and inhibitory conductances . The amplitude of the oscillatory component gIo is discussed in the Results . For a given input , MC activities were measured after at least one second of simulation to avoid transient effects . Euler integration was used with a time step of 0 . 02 ms . Simulations were performed using Matlab 7 . 0 ( The Mathworks , www . mathworks . com ) . | Olfactory function relies on a chain of neural relays that extends from the periphery to the central nervous system and implies neural activity with various timescales . A central question in neuroscience is how information is encoded by the neural activity . In the mammalian olfactory bulb , local neural activity oscillations in the 40–80 Hz range ( gamma ) may influence the timing of individual neuron activities such that olfactory information may be encoded in this way . In this study , we first characterize in vivo the detailed activity of individual neurons relative to the oscillation and find that , depending on their state , neurons can exhibit periodic activity patterns . We also find , at least qualitatively , a relation between this activity and a particular odor . This is reminiscent of general physical phenomena—the entrainment by an oscillation—and to verify this hypothesis , in a second phase , we build a biologically realistic model mimicking these in vivo conditions . Our model confirms quantitatively this hypothesis and reveals that entrainment is maximal in the gamma range . Taken together , our results suggest that the neuronal activity may be specifically formatted in time during the gamma oscillation in such a way that it could , at this stage , encode the odor . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] | [
"neuroscience/theoretical",
"neuroscience",
"biophysics/theory",
"and",
"simulation",
"neuroscience/sensory",
"systems",
"computational",
"biology/computational",
"neuroscience"
] | 2009 | Specific Entrainment of Mitral Cells during Gamma Oscillation in the Rat Olfactory Bulb |
The evolution of disease or the progress of recovery of a patient is a complex process , which depends on many factors . A quantitative description of this process in real-time by a single , clinically measurable parameter ( biomarker ) would be helpful for early , informed and targeted treatment . Organ transplantation is an eminent case in which the evolution of the post-operative clinical condition is highly dependent on the individual case . The quality of management and monitoring of patients after kidney transplant often determines the long-term outcome of the graft . Using NMR spectra of blood samples , taken at different time points from just before to a week after surgery , we have shown that a biomarker can be found that quantitatively monitors the evolution of a clinical condition . We demonstrate that this is possible if the dynamics of the process is considered explicitly: the biomarker is defined and determined as an optimal reaction coordinate that provides a quantitatively accurate description of the stochastic recovery dynamics . The method , originally developed for the analysis of protein folding dynamics , is rigorous , robust and general , i . e . , it can be applied in principle to analyze any type of biological dynamics . Such predictive biomarkers will promote improvement of long-term graft survival after renal transplantation , and have potentially unlimited applications as diagnostic tools .
Here we present a general framework that allows us to determine such an optimal coordinate or biomarker from longitudinal cohort studies directly , without constructing the Markov state model . The method was originally developed to describe complex dynamics of protein folding [7]–[9] . Briefly , a putative functional form of the reaction coordinate is assumed , for example , a linear combination of features ( here the metabolome NMR spectra ) that could describe the process . The approach is invariant to the choice of the functional form and the set of observables , provided they contain all the essential information about the dynamics of the process . The coordinate is optimized ( trained ) on a sample of trajectories , i . e . , realizations of a complex multidimensional dynamical process . This is achieved by choosing the coordinate ( e . g . , the coefficients of the linear combination ) such that the cut based free energy profile associated with the coordinate is the highest [7] , [10] . Namely , given an ensemble of N trajectories ( ) and a reaction coordinate functional form an ensemble of reaction coordinate trajectories is constructed by projecting the multidimensional trajectories onto the reaction coordinate ( ) . The optimal reaction coordinate is found by optimizing the parameters so that the cut-based free energy profile [7] , [8] is maximal . , where the partition function equals half number of transitions ( crossings ) by the reaction coordinate time-series through point y; here and below we set . The CFEP ( ) unlike the conventional histogram based free energy profile ( ) is invariant to reaction coordinate rescaling , insensitive to statistical noise and capable of detecting sub-diffusion . Together they determine the coordinate dependent diffusion coefficient and thus completely specify diffusive dynamics [9] . One can maximize instead the generalized cut based free energy profile where the partition function takes into account each transition through point y with weight equal to the transition distance; for a Gaussian distribution of steps ( i . e . , diffusive dynamics ) the two optimality criteria are equivalent [10] . If the reaction coordinate is a weighted sum of basis functions , as used here , the optimal values of the parameters ( that maximize can be found analytically [10] ) . In supervised optimization a coordinate that accurately describes the dynamics of transition between two given end states ( e . g . , healthy and disease ) is determined . Incidentally , the coordinate is the probability of full recovery , i . e . , of ending up in the “healthy” state rather than the “disease” state starting from a current state . It is known as committor or folding probability in protein folding studies [10] , [11] . If the two end states are separated by the highest free energy barrier , the transition between them corresponds to the slowest relaxation mode , and an eigenvector , corresponding to the slowest mode is an optimal reaction coordinate [10] , [12] . This coordinate can be determined in an unsupervised way without explicit definition of end states .
We have shown that the dynamics of recovery from kidney transplant can be quantitatively described as diffusion on a free energy profile , which is a function of a measurable biomarker . Such a biomarker can be determined in an ( un ) supervised way from longitudinal cohort studies ( patient trajectories ) , which is optimal in the sense that it is able to discriminate where each patient is on a free energy profile . In particular , the probability of rapid recovery ( primary function ) can be used to devise optimal treatment . Such an approach is general and can be useful to develop optimal biomarkers for diseases that develop slowly and in a complicated way depending on many factors , or unknown unknowns , such as aging [20] , cancer [21] , [22] and psychological disorders [23] .
Approval was given by the regional ethics approval committee approval number REC Ref: 07/H1306/129 The partition function of the cut-based free energy profile at point equals half the sum of the distances of those trajectory steps that go through point y [14] . More precisely , where ( x ) is the Heaviside step function and x ( iΔt ) is the reaction coordinate time series sampled with time interval Δt . The cut free energy profile is defined as and ; here we assume that . The optimal coordinate is defined as the one with the highest cut profiles ( lowest partition function ) . The justification of the optimization criteria can be summarized as follows ( for the details see the cited references ) . It can be shown that minimum of , with constrains and , is attained when the reaction coordinate y equals the coordinate - an optimal coordinate [10] . Correspondingly , a sub-optimal coordinate with a lower value of the cut profile , has the mean square displacement which grows slower then linear with time [14] . The latter is an indication that dynamics is not diffusive and that non-Markovain memory effects are at play . Another manifestation of a sub-optimal coordinate , is that it has lower free energy barriers and thus a faster kinetics . The kinetics along the coordinate with the highest cut profile is the slowest [9] , [14] . Optimization of the reaction coordinate can be performed in a supervised or unsupervised manner . In supervised optimization a coordinate that accurately describes the dynamics of transition between two given end states ( e . g . , healthy and disease ) is determined . Incidentally , the optimal coordinate is the probability of full recovery , i . e . , of ending up in the “healthy” state rather than the “disease” state starting from a current state . The optimization is constrained by fixing the value for the coordinate for the two state and [10] . If the reaction coordinate is a weighted sum of basis functions , boundary conditions are given as and where and index the points which belong to A and B states , respectively . The optimal weights ( ) which give constrained maximum to can be found analytically [10] . Here we specified the constrains in the following way , which resulted in a more flexible coordinate . Instead of assuming that each trajectory ends at either 0 or 1 , we assumed that each trajectory is constrained to end with either 0 or 1 ( in other words a trajectory reaches an end state on the following day ) . In this case the optimal weights are found by minimizing which equalswhere are indexes that refer to time frame , trajectory and basis function , respectively; the second term of the functional describes the boundary condition with equal 0 or 1 for trajectories connected to 0 or , respectively . The optimal parameters are found by solving the corresponding system of linear equations . To facilitate the visual comparison with the unsupervised results where changed to , which results in a shift and change of scale of the optimal coordinate . In unsupervised optimization the determined coordinate describes the slowest relaxation mode ( the second eigenvector ) of the stochastic dynamics [10] . If the two states ( A and B ) are separated by the highest barrier , so the slowest relaxation rate corresponds to the transition dynamics between the states , the second eigenvector reaction coordinate approximates the folding probability ( the probability of full recovery here ) reaction coordinate in the transition state region - the most important part for the description of the transition dynamics [10] , [12] . The eigenvectors can be found by minimizing under constraint [10] . Due to the constraint , the optimization function simplifies to the auto-correlation function . If reaction coordinate is a weighted sum of basis functions , the optimal weights can be found analytically . They are the solution of the generalized eigenvalue problem [10] . The free energy profile that describes the disease dynamics cannot be determined from the patients trajectory simply by computing the cut based ( or histogram ) free energy profiles because the trajectories are not at equilibrium . The procedure described in Ref [13] was employed . Briefly , assuming diffusive dynamics , the equilibrium free energy profile can be computed from the steady state ( non-equilibrium ) probability distribution asUsing , and , one obtainswhere and are the cut profiles that measure flux in positive and negative direction , respectively . Note that the method for determining the optimal reaction coordinate was originally derived for equilibrium dynamics; an extension of the framework to non-equilibrium dynamics has been suggested recently [24] . Here we assume that while non-equilibrium sampling affects populations , its main effect on the optimization procedure is in altering the contribution ( weight ) of the different regions to the optimization functional and can be neglected . The probability of “full recovery” ( the folding probability ) was computed from the free energy profile asThe success probability with 95% confidence interval were estimated from the trajectories by “add two successes and two failures” approach [25] as , where and and are the numbers of trajectories visiting bin ended up in left or right half of the profile , respectively [26] . NMR spectra were obtained for the water soluble components [27] of erythrocytes taken from 18 kidney transplant patients ( up to 9 time points from pre-op to 7 days after surgery ) . One-dimensional NMR spectra were measured at 499 . 97 MHz on a Varian Unity Inova 500 spectrometer at 20°C , using a standard PRESAT pulse sequence . For all samples a relaxation delay of ca . 9 s ( three times the longest T1 ) was applied between scans to allow the spins to fully relax , with 256 transients collected into 16384 data points and a spectral width of 6000 Hz . An exponential line broadening of 0 . 5 Hz was applied to each free induction decay ( FID ) and zero filling to 32768 points was carried out , followed by Fourier transformation . Phase and baseline corrections were carried out using ACD/Labs 12 . 0 ( Advanced Chemistry Development Inc . , Toronto , Canada ) and chemical shifts were referenced to the lactate doublet at 1 . 33 ppm . Independently of the NMR data , patients have been divided in three classes based on a clinical assessment of the patients into primary function ( PF ) , delayed graft function ( DGF ) and acute rejection ( AR ) with and without primary function . Primary function was defined as immediate recovery of renal function . Delayed graft function was defined as the need for dialysis in the first week following transplantation . Diagnosis of acute rejection was conducted on the basis of biopsy and histological findings . Dialysis was performed on day 5 to patient 4 , day 2 to patient 7 , day 4 to patient 9 , day 7 to patient 17 and day 6 to patient 18 . All biopsies were conducted between 6 and 9 days following transplantation . Nine patients had immediate primary function , five patients had delayed graft function and four patients had acute rejection . All but one transplant were eventually successful: in addition to acute rejection , patient 14 also suffered from renal artery stenosis , and the graft was ultimately removed . The immunosuppressive regime and induction agents were the same across the cohort . | The evolution of disease or the progress of recovery of a patient is usually monitored by collecting physical parameters , which may be simply the body temperature for a common cold or properties of tissue samples for e . g . , cancer . Most often clinical decisions are taken based on the current value or because of a sizable change of a relevant parameter . As more advanced diagnostic tools become available , and huge numbers of parameters can be collected at short , frequent time intervals , two related questions arise . The first is , which of the parameters provides relevant information on the progress of disease or recovery as opposed to noise ? Is there more information that can be obtained from the history of the evolution of such parameters ? Here we propose a novel approach that leads , for the specific case of recovery from kidney transplant , to a positive answer . | [
"Abstract",
"Introduction",
"Results/Discussion",
"Materials",
"and",
"Methods"
] | [
"physics",
"biochemistry",
"research",
"design",
"proteins",
"protein",
"folding",
"protein",
"structure",
"longitudinal",
"studies",
"biology",
"and",
"life",
"sciences",
"physical",
"sciences",
"biomarkers",
"biophysics",
"research",
"and",
"analysis",
"methods"
] | 2014 | Optimal Reaction Coordinate as a Biomarker for the Dynamics of Recovery from Kidney Transplant |
The Drosophila circadian oscillator controls daily rhythms in physiology , metabolism and behavior via transcriptional feedback loops . CLOCK-CYCLE ( CLK-CYC ) heterodimers initiate feedback loop function by binding E-box elements to activate per and tim transcription . PER-TIM heterodimers then accumulate , bind CLK-CYC to inhibit transcription , and are ultimately degraded to enable the next round of transcription . The timing of transcriptional events in this feedback loop coincide with , and are controlled by , rhythms in CLK-CYC binding to E-boxes . PER rhythmically binds CLK-CYC to initiate transcriptional repression , and subsequently promotes the removal of CLK-CYC from E-boxes . However , little is known about the mechanism by which CLK-CYC is removed from DNA . Previous studies demonstrated that the transcription repressor CLOCKWORK ORANGE ( CWO ) contributes to core feedback loop function by repressing per and tim transcription in cultured S2 cells and in flies . Here we show that CWO rhythmically binds E-boxes upstream of core clock genes in a reciprocal manner to CLK , thereby promoting PER-dependent removal of CLK-CYC from E-boxes , and maintaining repression until PER is degraded and CLK-CYC displaces CWO from E-boxes to initiate transcription . These results suggest a model in which CWO co-represses CLK-CYC transcriptional activity in conjunction with PER by competing for E-box binding once CLK-CYC-PER complexes have formed . Given that CWO orthologs DEC1 and DEC2 also target E-boxes bound by CLOCK-BMAL1 , a similar mechanism may operate in the mammalian clock .
Almost all organisms from Cyanobacteria to humans have internal circadian clocks that drive daily rhythms in physiology , metabolism and behavior , thereby synchronizing internal processes with the external environment . In eukaryotes , the circadian clock keeps time via one or more transcriptional feedback loops [1] . In Drosophila , a heterodimer formed by CLOCK ( CLK ) and CYCLE ( CYC ) binds E-box sequence activates transcription to initiate clock function . In the core loop , CLK-CYC activates period ( per ) and timeless ( tim ) transcription during mid-day , effecting a rise in per and tim mRNA levels that peaks during the early evening . PER and TIM proteins then accumulate , form a dimer , and move into the nucleus to bind CLK-CYC during the night , thereby inhibiting their transcriptional activity until PER and TIM are degraded early in the morning [2 , 3] . Another interlocked transcriptional feedback loop is also regulated by the core feedback loop . In this loop , CLK-CYC activates transcription of vrille ( vri ) and PAR-domain protein 1ɛ ( Pdp1ɛ ) , which bind D-boxes to repress and activate transcription , respectively , and drive RNA cycling of Clk and other output genes in the opposite phase as per , tim , vri and Pdp1ɛ [4–6] . PER was previously found inhibit CLK-CYC binding to E-boxes in vitro [7] , which suggests that the rhythmic transcription of CLK target genes are mediated by PER-dependent rhythms in E-box binding by CLK-CYC . Chromatin immunoprecipitation ( ChIP ) experiments using fly heads support this model , showing that CLK-CYC rhythmically bind E-boxes in the per circadian regulatory sequence ( CRS ) and the tim upstream sequence [8] . However , the mechanism by which CLK-CYC heterodimers are removed from E-boxes during repression is not well understood . PER is required for the rhythmic binding of CLK complexes , as CLK constantly binds to per and tim promoters in per01 flies [8] , indicating that PER inhibits transcription by removing CLK-CYC from E-boxes . Interestingly , co-expression of another transcription factor , CLOCKWORK ORANGE ( CWO ) , strongly enhanced PER-mediated repression in cultured Drosophila Schneider 2 ( S2 ) cells [9] , suggesting that PER is unable to efficiently remove CLK from DNA in the absence of other transcription repressors . Previous studies demonstrated that CWO , a basic helix-loop-helix ( bHLH ) -ORANGE transcriptional factor [10] , is a direct target of CLK-CYC [9 , 11 , 12] . In Drosophila Schneider 2 ( S2 ) cells , overexpression of CWO reduces the basal transcription of per , tim , vri and Pdp1ɛ promoter-driven luciferase reporter genes [9 , 12 , 13] . Furthermore , in the presence of PER , CWO repress CLK mediated transcription 5–10 fold in S2 cells , indicating that CWO is a strong transcription repressor that can cooperate with PER to repress CLK-CYC mediated transcription [9] . In cwo mutants or cwo RNAi knockdown flies , the levels of per , tim , vri and Pdp1ɛ mRNAs are increased during the early to mid-morning [9 , 12] . These results suggest that CWO co-represses CLK-CYC activity along with PER during the end of a cycle [9 , 12] . However , the mechanism through which CWO represses CLK-CYC mediated gene transcription remains unknown . In this study we demonstrate that CWO and CLK bind core clock gene E-boxes in a reciprocal pattern across the circadian cycle in vivo , which suggests that CWO competes with CLK to bind E-boxes . We also show that CWO acts to decrease CLK binding to tim E-boxes during early morning , when PER binds CLK-CYC to reduce its binding to DNA [8] , but not during early night when CLK-CYC strongly binds E-boxes in the absence of PER . These results suggest a model for CWO function where CWO has low DNA binding affinity compared to CLK-CYC complexes during the activation phase , but has higher affinity compared to CLK-CYC-PER complexes , and is thus capable of removing CLK-CYC-PER complexes from E-boxes to consolidate and maintain repression . Constant high CWO binding to the tim promoter in Clkout flies ( i . e . comparable to binding at ZT2 in wild-type ) and constant low CWO binding in per01 flies ( i . e . comparable to binding at ZT14 in wild-type ) supports our model for CWO repression . As a whole , these results suggest that CWO co-represses CLK-CYC activity with PER by competing with CLK-CYC-PER complexes for E-box binding , therefore promoting the transition to off-DNA repression .
Earlier studies demonstrated that cwo mRNA cycles in phase with per , tim , vri , and Pdp1 , but with a higher basal level , and thus lower amplitude [9 , 12–14] . To determine whether CWO protein levels also cycle , western analysis was carried out using head extracts from wild-type flies collected every 4 hours in a 12-h light/12-h dark ( LD ) cycle . We find that the levels of CWO do not change throughout an LD cycle ( Fig 1 ) , consistent with previous results [15] . Given that cwo mRNA levels cycle , it is possible that constant CWO levels result from post-transcriptional regulation or a long half-life . CWO contains a bHLH domain , a structural motif that characterizes a family of E-box binding transcription factors [16–19] , which suggests that CWO may regulate CLK-CYC target gene transcription via E-box binding . Previous ChIP-on-chip and gel-shift analyses in S2 cells demonstrated that CWO specifically binds to the E-box of core clock genes [12 , 13] , however it is still unknown whether CWO binds those core clock genes in vivo , and whether the binding intensity changes throughout the day . To test these possibilities , ChIP assays were carried out on wild-type flies collected in the early morning ( ZT2 ) and in the early night ( ZT14 ) using CWO and CLK antisera . Fragments containing upstream E-boxes from tim , per , Pdp1 and vri , which are necessary for high-amplitude mRNA cycling in vitro or in vivo [4 , 5 , 20–25] , were amplified from the immunoprecipitates and then quantified . In CWO immunoprecipitates , the tim , vri and Pdp1 E-box containing fragments were two to threefold more abundant at ZT2 than at ZT14 ( Fig 2A ) , suggesting that CWO binding is time-dependent , though the dynamic binding of CWO on the per E-box fragment is less robust than the others . Importantly , this temporal binding pattern is antiphase to CLK binding , as CLK shows high binding intensity during the night at ZT14 and low binding during the daytime at ZT2 ( Fig 2B ) , consistent with previous results [8 , 11] . The reciprocal binding pattern of CLK and CWO implies that these transcription factors compete for E-box binding . If so , both CLK and CWO must occupy the same E-boxes . To test this possibility , we determined how mutating E-boxes upstream of tim affected CLK and CWO binding . The circadian enhancer upstream of tim is comprised of two tandem E-boxes that are spaced seven nucleotides apart [24 , 26] , a structure that is conserved among core clock genes in various species [27] . Both of these E-boxes were indispensable for tim mRNA expression in S2 cells [24] , suggesting that these tandem E-box motifs are binding sites for both CLK and CWO . To determine if this is the case , a series of 136bp fragments from the tim promoter containing an E-box1 ( E1 ) mutant ( mE1-E2 ) , an E-box 2 ( E2 ) mutant ( E1-mE2 ) , an E1 and E2 double mutant ( mE1-mE2 ) or a control with wild-type E-boxes ( E1-E2 ) were generated , inserted into the pHPdestGFP vector [28] , and targeted to the attP18 genomic site ( Fig 3A ) . To confirm that this promoter fragment is sufficient to drive rhythmic expression , we carried out quantitative reverse transcription-PCR ( qRT-PCR ) to monitor GFP mRNA levels in flies collected every 4-h during an LD cycle . Quantification of GFP mRNA levels in flies with WT tim promoter shows a ~10-fold diurnal rhythm with a peak at ZT14 and a trough at ZT2 to ZT6 ( S1 Fig ) , consistent with timing and amplitude of per and tim mRNA cycling in wild-type flies [29 , 30] . However , even at the normal tim mRNA peak ( ZT14 ) , mE1-E2 , E1-mE2 and mE1-mE2 flies express little or no eGFP mRNA ( S1 Fig ) , indicating that both E1 and E2 are indispensable for expression of tim mRNA in vivo . This result is consistent with previous tim-luciferase reporter results in S2 cells [24] . We next carried out ChIP assays using CWO and CLK antisera on the same fly strains to test whether E1 and E2 are required for CWO and CLK binding . At ZT2 , when CWO strongly binds to the tim promoter , CWO binding intensity was drastically reduced in mE1-E2 , E1-mE2 and mE1-mE2 flies compared to WT ( Fig 3B ) . Likewise , CLK binding intensity was drastically reduced in mE1-E2 , E1-mE2 and mE1-mE2 flies compared to WT at ZT14 , when CLK binding is strongest ( Fig 3B ) . These results indicate that both E1 and E2 are indispensable for both CWO and CLK binding to the tim circadian enhancer . Given that CWO specifically targets E-boxes in S2 cells by Gel-shift analyses [13] , we conclude that both CLK and CWO bind intact tandem E1-E2 motifs in vivo . In mice , CLK-BMAL1 dimers cooperatively bind tandem E-boxes in vitro [27 , 31] , and this may be the case for CWO given the requirement for both E1 and E2 E-boxes . Previous studies showed that increasing the level of CWO expression reduces per , tim , vri and Pdp1ɛ mRNA levels in S2 cells and that their trough mRNA levels are higher in cwo mutant or knockdown flies , indicating that CWO acts to repress CLK-mediated gene transcription in vitro and in vivo [9 , 12–14] . Given that CWO and CLK bind to the same E-box motif , we wondered whether CWO represses CLK-mediated transcription by inhibiting CLK binding . To test this possibility , ChIP assays were carried out using CLK antiserum on wild-type and cwo5703 flies at the trough ( ZT2 ) and peak ( ZT14 ) times of CLK-CYC target gene transcription and mRNA abundance in LD . Although cwo5703 mutants lengthen the period of activity rhythms by 2–3h in DD [9 , 13] , the peak and trough phases of CLK-CYC target gene transcription and mRNA abundance are comparable in cwo5703 mutants and wild-type flies in LD [9 , 13] . We find that CLK binds tim E-boxes with a robust rhythm in wild-type flies and a lower amplitude rhythm in the cwo5703 mutant ( Fig 4A ) . However , the intensity of CLK binding in cwo5703 is significantly increased at ZT2 compared to wild-type , indicating that CWO acts to reduce CLK-CYC binding at the trough of its binding cycle ( Fig 4B ) . Given that CWO strongly binds tim E-boxes at ZT2 ( Fig 2A ) , we propose that CWO inhibits CLK-CYC binding during the repression phase by antagonizing PER-CLK-CYC complexes to maintain off-DNA repression . There was no significant difference in CLK binding between cwo5703 and wild-type at ZT14 ( Fig 4B ) , despite decreased peak levels of per , tim , vri and Pdp1ɛ mRNA at ZT14 in cwo mutant and RNAi knockdown flies [9 , 12–14] , suggesting that CWO has little impact on CLK-CYC binding in the absence of PER . Given that CWO suppresses CLK binding at ZT2 in the early morning but not at ZT14 during the early evening ( Fig 4B ) , it is possible that PER is necessary for CWO to antagonize CLK E-box binding since PER accumulates to high levels in the nucleus around dawn and is at low levels in the cytoplasm around dusk [32] . Indeed , our results support a model developed previously to explain cooperation between CWO and PER to repress CLK-CYC mediated transcription in S2 cells [9] . In this model , CWO is proposed to compete with CLK-CYC heterodimers for E-box binding only when PER binds CLK-CYC , thereby reducing their affinity for E-box binding . To test this model , we performed ChIP assays using CWO antiserum on wild-type , Clkout and per01 flies collected at ZT2 and ZT14 in LD . In Clkout flies , which necessarily lack CLK-CYC heterodimers [33] , CWO is bound to tim E-boxes at both ZT2 and ZT14 with binding signals comparable to the strong CWO binding in wild-type flies at ZT14 ( Fig 5A ) . In contrast , in per01 flies , which lack PER-dependent repression of CLK-CYC activation [34] , low binding signals of CWO were detected at ZT2 and ZT14 , indicating that PER is indeed required for CWO to bind E-boxes ( Fig 5A ) . Moreover , CWO binding was significantly increased in Clkout versus wild-type flies at ZT14 , indicating that CLK-CYC binding at ZT14 reduces CWO binding . Likewise , a significant increase in CWO binding was also seen in wild-type versus per01 flies at ZT2 , indicating that PER enhances CWO binding ( Fig 5A ) . To determine whether differences in CWO binding in Clkout and per01 flies were due to differences in CWO protein levels , we carried out western analysis using head extracts from these mutants collected at ZT2 and ZT14 . Since cwo transcription is regulated in part by the transcriptional feedback loop , CWO protein levels are slightly lower in Clkout flies and slightly higher in per01 flies ( Fig 5B and 5C ) . However , the lower levels of CWO in Clkout resulted in higher E-box binding , and higher CWO protein levels in per01 resulted in lower E-box binding . This result suggests that the differences in CWO-E-box binding are not due to altered CWO protein levels , but due to the relative DNA binding affinities of CWO and CLK in these mutants . These results , taken together , strongly support and extend the model described by Kadener et al . , 2007 , for CWO binding as it relates to CLK-CYC repression . When CLK-CYC targets are activated , CLK-CYC binds DNA with higher affinity than CWO , thus CLK binding is not altered in the presence or absence of CWO . When CLK-CYC targets are repressed , PER binds CLK-CYC complexes and decreases their DNA binding affinity , thereby favoring CWO binding to E-boxes and enhancing PER mediated removal of CLK-CYC-PER complexes from the DNA ( Fig 6 ) . Although we can’t exclude the possibility that PER enables CWO E-box binding independent of its interaction with CLK-CYC , the available evidence strongly supports the model proposed .
Rhythmic binding of CLK-CYC to E-boxes is essential for rhythmic transcription of the core circadian oscillator genes per and tim in Drosophila . CLK-CYC bind E-boxes upstream of per and tim in the late day and early night to activate transcription; and is released from these binding sites during late night [8 , 35 , 36] . Previous work demonstrated that CLK constitutively binds per and tim E-boxes in per01 flies , indicating that PER is essential for rhythmic binding of CLK-CYC , and is key to removing CLK-CYC from E-boxes [8] . In this study we report that CWO also contributes to removing CLK-CYC from E-boxes . In cwo5703 mutant flies , CLK binding intensity is significantly increased at the trough of its binding cycle , suggesting that repression is incomplete in the absence of CWO ( Fig 4 ) . We find that CWO and CLK bind E-boxes upstream of tim in a reciprocal manner during a daily cycle , and that CLK shows significantly increased binding intensity at the trough of its binding cycle in cwo mutant flies , indicating that CWO acts to antagonize CLK-CYC binding . Given that both CWO and CLK are constitutively expressed ( Fig 1; [8] ) , we believe that the key driver for the transition between dynamic CLK-CYC and CWO binding is the accumulation of PER , which alters the relative affinity of E-box binding by CLK-CYC . CWO shows low levels of tim E-box binding in per01 flies , in which CLK-CYC constantly bind E-boxes , but shows high levels of tim E-box binding in Clkout flies that lack CLK expression and E-box occupancy . These results suggest that CWO E-box binding affinity is lower than the CLK-CYC heterodimer and higher than the CLK-CYC-PER complex , which could account for the PER-dependent rhythms in CLK-CYC and CWO binding ( Fig 6 ) . During late day and early night , CLK-CYC binds E-boxes to activate transcription in the presence of CWO because CLK-CYC has higher DNA binding affinity . PER starts to accumulate in the nucleus during the night and interacts with CLK-CYC , which decreases CLK-CYC DNA interaction via reduced DNA binding affinity . Consequently , CWO displaces CLK-CYC-PER from E-boxes by binding with comparatively higher affinity . Once CLK-CYC-PER is removed , CWO occupancy on E-boxes prohibits CLK-CYC-PER from re-binding , thus maintaining transcriptional repression ( Fig 6 ) . Unlike the constitutive CLK-CYC E-box binding in per01 flies [8] , CLK-CYC binding is rhythmic in cwo5703 flies , but with a dampened amplitude due to elevated CLK binding at the trough ( Fig 4A ) . This low amplitude rhythm in CLK binding may explain why a large proportion of cwo5703 flies show long period rhythms rather than losing rhythmicity entirely like per01 mutants [9 , 12–14] . We speculate that the long period phenotype is caused in part by a prolonged repression process . Based on the current model for repression of CLK-CYC transcription , PER-TIM complexes first bind CLK-CYC , thereby removing CLK-CYC from the E-boxes and inhibiting per and tim transcription , then PER and TIM degradation enables CLK-CYC binding to start another cycle of transcription [3] . Both of these steps could be delayed in a cwo mutant . In the absence of CWO it takes longer to remove CLK-CYC from the DNA; PER alone can repress CLK-CYC binding to some degree , but CLK-CYC-PER complexes still weakly bind E-boxes if CWO is absent , thus reducing CLK-CYC repression compared to wild-type flies . The outcome of incomplete repression of CLK-CYC E-box binding would be an increase in the trough levels of per and tim mRNAs , which is exactly what was observed in cwo mutant and RNAi knockdown strains [9 , 12–14] . Higher per and tim mRNA levels would in turn increase PER and TIM expression during the repression phase [14] . Higher levels of PER and TIM would not repress CLK-CYC binding efficiently in the absence of CWO , but would take longer to be degraded , thereby delaying the next cycle of transcriptional activation . In addition to the increased trough levels of core clock gene mRNAs in cwo mutant and RNAi knockdown flies , the peak levels of these mRNAs are lower , particularly during DD [9 , 12–14] . Decreasing per mRNA levels also lengthen circadian period [37] , thus making it difficult to determine the extent to which a lower mRNA peak or increased mRNA trough contributes to period lengthening in cwo mutant and RNAi knockdown flies . CLK binding at the peak of transcription is not significantly lower in cwo5073 than wild-type during LD ( Fig 4B ) , which argues that CWO enhances CLK-CYC transcriptional activity independent of CLK-CYC E-box binding . Additional experiments will be needed to decipher the mechanism underlying this CWO dependent increase in CLK-CYC transcription . PER dependent repression of CLK-CYC transcription is thought to occur in two stages . First , PER is recruited to circadian promoters by interacting with CLK to form PER-CLK-CYC complexes “on-DNA” , which inhibit CLK-CYC dependent transcription via an unknown mechanism . Subsequently , a decrease in the DNA binding affinity of PER-CLK-CYC complexes results in their release from DNA to initiate ‘‘off-DNA” phase of repression [35] . According to our model , CWO is critical for the transition to , and maintenance of , off-DNA repression . When PER-CLK-CYC complexes with low DNA affinity are formed , CWO promotes off-DNA repression by competing with CLK-CYC-PER complex for E-box binding . CWO occupancy on E-boxes then prevents PER-CLK-CYC from re-binding , thereby maintaining off-DNA repression . In mammals , a similar pattern of antagonistic binding on E-boxes between transcription factors was recently reported; USF1 and a mutant form of CLOCK , CLOCKΔ19 , bind to the same tandem E-boxes in a reciprocal manner . Wild-type CLOCK-BMAL1 complex binds E-boxes with much higher affinity than USF1 , but CLOCKΔ19-BMAL1 binds E-boxes with a similar affinity to USF1 , thus allowing USF1 to bind E-boxes [31] . Although this competitive binding is not thought to impact feedback loop function under normal circumstances , it demonstrates that other transcription factors can out-compete CLOCK-BMAL1 for E-box binding if the DNA binding affinity of CLOCK-BMAL1 is reduced . In this case CLOCK-BMAL1 binding is compromised by the ClockΔ19 mutation , but other mechanisms such as interactions with repressors and protein modifications could also reduce the binding affinity of CLOCK-BMAL1 or its orthologs . As in Drosophila , rhythmic binding of CLOCK-BMAL1 to E-boxes drives circadian transcription in mammals ( reviewed in [38] ) . Recent ChIP-seq analyses in mouse liver revealed time-dependent binding of CLOCK , BMAL1 and key negative feedback components including PER1 , PER2 , CRY1 and CRY2 [27 , 39–41] . The mechanism underlying the dynamic DNA occupancy of these transcription factors is not known , but previous work shows that the PER2-CLOCK interaction is required to initiate repression of CLOCK-BMAL1 dependent transcription [42] , which suggests that CLOCK-BMAL1 may be removed from E-boxes by the same mechanism as CLK-CYC in Drosophila . A recent genome-wide nucleosome analysis in mouse liver revealed that rhythmic E-box binding by CLOCK-BMAL1 removes nucleosomes [43] . However , despite rhythmic CLOCK-BMAL1 binding , nucleosome occupancy on E-boxes is always well below surrounding sequences , even in Bmal1-/- mutant livers [43] . This result indicates that chromatin at CLOCK-BMAL1 target sites is not closed even when there is no CLOCK-BMAL1 binding , suggesting that other transcription factors may occupy these E-boxes when CLOCK-BMAL1 is absent . These results , taken together , suggest that rhythms in activator binding may be controlled by a common mechanism in Drosophila and mammals . The mammalian orthologs of CWO , called DEC1 and DEC2 ( and also SHARP2 and SHARP1 , respectively ) , suppress CLOCK-BMAL1-induced activation [44–50] . Gel mobility shift and ChIP assays in vitro revealed that both DEC1 and DEC2 bind to E-box motifs targeted by CLK-BMAL1 [45–49] , and the DNA-binding domain is required for DEC1 to regulate CLK-BMAL1-induced transactivation [48] . In addition , DEC1/2 shows synergistic activity to PER1 in the regulation of clock gene mRNA levels in the SCN , as exemplified by significant changes in the period of circadian activity rhythms when null mutants for Dec1 , Dec2 or both Dec1 and Dec2 are combined with that for Per1 [44] . In contrast to the constant levels of CWO , DEC1 protein is rhythmically expressed in mouse liver , where DEC1 levels are high when PER-CRY complexes repress CLK-BLMAL1 transcription [51] . Taken together , these results raise the possibility that DEC1 and DEC2 may be a functional counterpart of CWO in competing with CLOCK-BMAL1 for E-box binding to repress CLOCK-BMAL1-mediated transcription .
DNA fragments containing wild-type or mutant E-boxes from the upstream tim circadian enhancer were used to construct GFP-reporter transgenes . These 136bp fragments extend from -578 to -714 relative to the tim transcription start site , and contain “E1-E2” E-box motifs that are wild-type ( E1-E2 ) , E1 mutant ( mE1-E2 ) , E2 mutant ( E1-mE2 ) or E1-E2 mutant ( mE1-mE2 ) . These wild-type and mutant E-box fragments were generated by PCR amplification using the following primer sets: E1-E2 , 5’-CACCTTTGGCAAATAAACGTGCGGCA-3’ and 5’-TGCCGGCGTTTGTGCGAA-3’; mE1-E2 , 5’-CACCTTTGGCAAATAAACGTGCGGCACGTTGTGATTAAGATCTAGCCGAT-3’ and 5’-TGCCGGCGTTTGTGCGAA-3’; E1-mE2 , 5’-CACCTTTGGCAAATAAGATCTCGGAGATTTGTGATTACACGTGAGCCGAT-3’ and 5’-TGCCGGCGTTTGTGCGAA-3’; mE1-mE2 , 5’-CACCTTTGGCAAATAAGATCTCGGAGATTTGTGATTAAGATCTAGCCGAT-3’ and 5’-TGCCGGCGTTTGTGCGAA-3’ . The PCR products were inserted into the pENTR/D-TOPO vector using pENTR Directional TOPO cloning kit ( Invitrogen ) , and then subcloned into the pHPdesteGFP vector , which expresses Green Fluorescent Protein ( GFP ) according to the enhancer sequence inserted [28] , using Gateway LR-Clonase System ( Invitrogen ) . The nucleotide sequences of all transgenes were confirmed by sequencing . The resulting transgenes were injected into embryos ( BestGene ) for recombination into the attp18 genomic site via PhiC31-mediated transgenesis to yield tim circadian enhancer GFP ( tim-CEG ) flies [52–54] . Flies were entrained in a 12-h light/12-h dark ( LD ) incubator for at least 3 days , collected at the indicated time points , and frozen . Isolated frozen fly heads were homogenized in radioimmunoprecipitation assay ( RIPA ) buffer ( 20 mM Tris at pH 7 . 5 , 150 mM NaCl , 1 mM EDTA , 0 . 05 mM EGTA , 10% glycerol , 1% Triton X-100 , 0 . 4% sodium deoxycholate ) containing 0 . 5 mM PMSF ( phenylmethylsulfonyl fluoride ) , 10 μg/ml aprotinin , 10 μg/ml leupeptin , 2 μg/ml pepstatin A , 1 mM Na3VO4 , and 1 mM NaF . This homogenate was sonicated 3 to 5 times for 10 s each time , using a Misonix XL2000 model sonicator at a setting of 3 and then centrifuged at 20 , 000 g for 10 min . The supernatant was collected as RIPA S extract , and protein concentration was determined by the Bradford assay . Equal amounts of RIPA S extract were run , transferred , and probed with guinea pig anti-CWO ( GP-27 ) , 1:5 , 000 and mouse anti-beta-actin ( Abcom ) , 1:20 , 000 . Horseradish peroxidase-conjugated secondary antibodies ( Sigma ) against guinea pig and mouse were diluted 1:5 , 000 . Immunoblots were visualized using ECL plus ( GE ) reagent . Protein levels were measured by placing a rectangle of the same size over each CWO , ß-Actin or non-specific ( NS ) protein band on films used to visualize the immunoblots , and quantifying the signal within each rectangle via densitometric analysis using the ImageJ program . The levels of CWO were calculated as a CWO:ß-Actin or CWO:NS ratio , and CWO abundance at each time point was plotted relative to wild-type at ZT2 . Chromatin IP ( ChIP ) assays and qPCR quantification were performed as previously described [55] . CLK and CWO binding to E-boxes in the circadian enhancers upstream of tim , per , vri , and Pdp1 in wild-type flies and the circadian enhancer in tim-CEG flies were first quantified via qPCR , and the resulting values were corrected for nonspecific binding to an intergenic region on chromosome 3R ( nucleotides 29576172 to 29576303 ) . The primers used for qPCR were as follows: for tim E-boxes , 5’-ACACTGACCGAAACACCCACTC-3’ and 5’-GCGGCACGTTGTGATTACACG-3’; for per E-boxes , 5’-GGGTGAGTAATGCCGTTGCGAAAT-3’ and 5’-ATTTGCTGGCCAAGTCACGCAGTT-3’; for vri E-boxes , 5’-CTGGTGCCTCACATTCCACG-3’ and 5’- CAGCAGTCAAGTTATAGCAGCGC-3’; for Pdp1 E-boxes , 5’-GCACTCTCATTCTCTCTGTCGC-3’ and 5’-ACTTGGGGGACTGGAACTG-3’; for tim-CEG , 5’-GCCCCCTTCACCTTTGGCAAATA-3’ and 5’-TACAAGAAAGCTGGGTCGGCG-3’; and for the intergenic region , 5’-CAGGAGTCGVAGGACCAACC-3’ and 5’-GTCCTGAGAGGCTGAGAGGC-3’ . PCR amplification using each pair of primers produced a single band of the expected size . The tim-CEG primers target vector sequences that flank the genomic tim E-box insert , and thus do not amplify endogenous tim genomic sequences . Quantitative RT-PCR was performed as described [55 , 56] , with some modifications , to measure GFP mRNA levels . Total RNA was isolated from frozen fly heads using Trizol ( Invitrogen ) , and treated with a Turbo DNase DNA-free kit ( Ambion ) to eliminate genomic DNA contamination . DNA-free total RNA ( 1 . 0 μg ) was reverse transcribed using oligo ( dT ) 12–28 primers ( Invitrogen ) and Superscript II ( Invitrogen ) . The reverse transcription ( RT ) product was amplified with SsoFast qPCR Supermix ( Bio-Rad ) in a Bio-Rad CFX96 Real-Time PCR System using primers to GFP ( 5’-TACGGCAAGCTGACCCTGAAGT-3’ and 5’-CGCACCATCTTCTTCAAGGACG-3’ ) and ribosomal protein 49 ( rp49 ) ( 5’-TACAGGCCCAAGATCGTGAA-3’ and 5’-GCACTCTGTTGTCGATACCC-3’ ) . For each sample , mRNA quantity was determined using the standard curve for each gene analyzed . To determine the relative levels of GFP mRNA over a diurnal cycle , GFP mRNA levels were divided by rp49 mRNA levels for each time point and plotted as the GFP/rp49 mRNA ratio . To quantify GFP mRNA in different tim-CEG strains at the wild-type ( E1-E2 ) peak , GFP/rp49 values were normalized to the E1-E2 value at ZT14 . | Circadian clocks control daily rhythms in animal , plant and fungal physiology , metabolism and behavior via transcriptional feedback loops . In Drosophila , the CLOCK-CYCLE ( CLK-CYC ) activator complex binds E-box regulatory sequences to initiate transcription of hundreds of effector genes including their own repressors , PERIOD ( PER ) and TIMELESS ( TIM ) , which feed back to repress CLK-CYC until they are degraded , thus allowing another cycle of CLK-CYC activation . Although the repression process is critical for the stability and accuracy of circadian timekeeping , how PER-TIM complexes maintain a transcriptionally repressed state for many hours is not well understood . Here we demonstrate that the transcription factor CLOCKWORK ORANGE ( CWO ) antagonizes CLK-CYC E-box binding , thus enhancing the removal of CLK-CYC from E-boxes to maintain transcriptional repression . This process requires PER , which suggests that PER-TIM and CWO cooperate to maintain a transcriptionally repressed state by removing CLK-CYC from E-boxes . These results demonstrate that PER-TIM requires CWO to effectively repress circadian transcription , and given that circadian transcriptional regulators are well conserved , this mechanism may function to repress transcription in other animals including humans . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"invertebrates",
"gene",
"regulation",
"regulatory",
"proteins",
"messenger",
"rna",
"dna-binding",
"proteins",
"animals",
"dna",
"transcription",
"circadian",
"oscillators",
"animal",
"models",
"drosophila",
"melanogaster",
"model",
"organisms",
"transcription",
"factors",
"chronobiology",
"drosophila",
"research",
"and",
"analysis",
"methods",
"proteins",
"gene",
"expression",
"insects",
"arthropoda",
"biochemistry",
"rna",
"circadian",
"rhythms",
"nucleic",
"acids",
"genetics",
"biology",
"and",
"life",
"sciences",
"organisms"
] | 2016 | CLOCKWORK ORANGE Enhances PERIOD Mediated Rhythms in Transcriptional Repression by Antagonizing E-box Binding by CLOCK-CYCLE |
Buruli ulcer ( BU ) is a necrotizing bacterial infection of skin , subcutaneous tissue and bone caused by Mycobacterium ulcerans . Although the functional impairment caused by BU results in severe suffering and in socio-economic problems , the disease remains largely neglected in Africa . The province of Bas-Congo in Democratic Republic of Congo contains one of the most important BU foci of the country , i . e . the Songololo Territory in the District of Cataractes . This study aims to assess the impact of a BU control project launched in 2004 in the Songololo Territory . We used a comparative non-randomized study design , comparing clinical profiles and outcomes of the group of patients admitted at the General Reference Hospital ( GRH ) of the “Institut Médical Evangélique” ( IME ) of Kimpese 3 years before the start of the project ( 2002–2004 ) with those admitted during the 3 years after the start of the project ( 2005–2007 ) . The BU control project was associated with a strong increase in the number of admitted BU cases at the GRH of IME/Kimpese and a fundamental change in the profile of those patients; more female patients presented with BU , the proportion of relapse cases amongst all admissions reduced , the proportion of early lesions and simple ulcerative forms increased , more patients healed without complications and the case fatality rate decreased substantially . The median duration since the onset of first symptoms however remained high , as well as the proportion of patients with osteomyelitis or limitations of joint movement , suggesting that the diagnostic delay remains substantial . Implementing a specialized program for BU may be effective in improving clinical profiles and outcomes in BU . Despite these encouraging results , our study highlights the need of considering new strategies to better improve BU control in a low resources setting .
Buruli ulcer ( BU ) is a necrotizing bacterial infection of skin , subcutaneous tissue and bone , caused by an environmental pathogen , Mycobacterium ulcerans [1] . Although the functional impairment caused by BU results in severe suffering and in socio-economic problems [2] , the disease remains largely neglected by health authorities in Africa [3] . BU is considered as one of the Neglected Tropical Diseases with a poorly known global prevalence [4] . The province of Bas-Congo ( Lower Congo ) in the Democratic Republic of Congo ( DRC ) contains one of the most important BU foci of the country , i . e . the Songololo Territory in the District of Cataractes [5]–[10] . Meyers et al . reported that BU existed in that region before 1935 on the basis of interviews of former patients [7] . The first BU case reports were published in the sixties [5]–[7] followed by a long period without reported cases . Since 1999 , the general reference hospital ( GRH ) of the Institut Médical Evangélique ( IME ) /Kimpese , located in the Songololo Territory , 220 km southwest of Kinshasa , regularly admits BU cases . Between 2002 and 2004 this hospital admitted 64 patients , 95% of them in the ulcerative stage . During this period , 48 patients out of 64 ( 75% ) were referred by government health centers or other health professionals , 9 ( 14 . 1% ) by family members , and 7 ( 10 . 9% ) presented spontaneously . Surgery was the main method of treatment applied amongst these patients ( 93 . 7% ) . An abnormally high case fatality rate ( 18 . 7% ) was observed among these 64 patients , and whereas 36% presented already a functional limitation at the time of diagnosis , 23% were discharged with permanent disability . The median length of hospitalization was 89 days and , -noteworthy- 90% of the patients were not able to pay their hospitalization costs . To address these poor clinical outcomes , the American Leprosy Mission and the IME hospital launched a BU control project in Songololo Territory in 2004 . The principal aims of this project were ( i ) the improvement of the patient care of BU patients admitted at the GRH IME/Kimpese and ( ii ) the promotion of early community-based detection of suspected BU cases . The aim of this study is to evaluate the impact of this specialized BU control program on clinical profiles and outcomes .
Ethical clearance for this study was obtained from the Institutional Review Board of IME . All patients , or their guardian in the case of minors , provided informed consent for all diagnostic and treatment procedures and publication of any or all images derived from the management of the patient , including clinical photographs that might reveal patient identity . The BU control project started at the end of 2004 and introduced free patient care for BU patients during their admission at GRH IME/Kimpese , whereas this was hitherto to be paid on a fee-for-service basis . Furthermore , the patients benefited from a free daily nutritional supplement , and specific antibiotherapy was introduced in accordance with WHO recommendations [11] , as well as a physiotherapy program for prevention of disabilities . Simultaneously the project organized awareness raising campaigns in the endemic communities , based on a mass-media approach targeting the general public , followed by active case-finding and referral of suspected cases to the specialized BU care centre . The project was based on the following five components: Improving facilities' management and treatment skills; Prevention of disabilities and physical rehabilitation; Feeding patients and psychological and social support for those affected; Stepping up Information , Education and Communication for the general public and community-based surveillance , and Training and research . To evaluate the effect of this control project , we used a comparative non-randomized study design , comparing patient demographic profiles and clinical outcomes of the group of patients admitted at the GRH IME/Kimpese in the 3 years before the start of the project ( 2002–2004 ) with those admitted during the 3 years after the start of the project ( 2005–2007 ) . We have included all consecutive patients clinically diagnosed as BU and admitted to the Surgical Department of GRH IME/Kimpese from January 2002 to December 2007 . The clinical case definition elaborated by the World Health Organisation ( WHO ) was used to diagnose BU [12] . Additionally for the second period , as recommended by the WHO [11] , we introduced patients' categorization as follows: A single lesion <5 cm ( Category I ) ; A single lesion 5–15 cm ( Category II ) ; A single lesion >15 cm , multiple lesions , lesions at critical sites ( face , breast and genitalia ) or osteomyelitis ( Category III ) . For all patients included in this study , the diagnostic confirmation process consisted of swabs from ulcerative lesions and biopsies for the laboratory confirmation ( bacteriology and/or histopathology ) of suspected cases according to WHO recommendations [12] . The initial direct smear examinations for acid-fast bacilli and histopathologic analyses were made at the IME/Kimpese laboratory . Other specimens from the same patient were sent in a transport medium to the Mycobacteriology Unit of the Institute of Tropical Medicine ( ITM ) in Antwerp , Belgium [13] , where Ziehl-Neelsen ( ZN ) staining , in vitro culture on Löwenstein-Jensen medium , and PCR for the detection of M . ulcerans DNA were performed according to WHO recommendations [12] . Formalin-fixed tissues were sent to the Department of Infectious and Parasitic Diseases Pathology of the Armed Forces Institute of Pathology in Washington DC , for the histopathological confirmation of diagnosis [10] . Throughout the whole study period , clinical data of BU patients were recorded on standardized Case Report Forms elaborated by WHO ( known as form BU01 ) and the data were entered in a standardized case registry form ( BU02 ) [14] . Next , these data were entered into an Excel database ( Microsoft Corporation , Redmond , WA ) and analyzed with Epi-Info version 3 . 3 . 2 ( Centers for Diseases Control and Prevention , Atlanta , GA ) . The Pearson chi-square test was used to compare proportions with a significance level set at 5% , as well the Fisher's exact test when an expected cell value was less than 5 . To evaluate the relevance and the effect of the BU control project , we used the conceptual framework to evaluate public health programs proposed by Habicht et al . [15] . The principal indicators considered for the data analysis are the number of recorded cases for each period , the number of new cases and relapses , the proportion of cases with functional limitation of joints at diagnosis , the proportion of cases confirmed by at least one laboratory test , the proportion of ulcerative forms at diagnosis , the type of treatment applied , the proportion of discharged cases with functional limitation of joints , the median duration of hospitalization , and the case-fatality rate . Relapse was defined in both study periods as a new confirmed diagnosis of BU less than one year after being declared cured from BU after treatment ( surgical only in the first period , antibiotic and/or surgical in the second period ) . Functional limitation was defined as any reduction in the range of motion of one or more joints , and was assessed based on clinical observation . Lesions were considered as mixed forms when simultaneous presence of different forms of disease including bone and joint involvement in the same patient was noticed . Besides , we defined as simple ulcerative forms ( SUF ) the ulcerative lesions not associated with other clinical lesions such as papule , nodule , plaque , edema or osteomyelitis at the same site .
The number of suspected BU cases admitted at GRH IME/Kimpese strongly increased after the start of the BU control project . The average number of annual admissions for BU tripled , from 21 cases per year for the period 2002–2004 , to 63 cases per year for 2005–2007 ( Figure 1 ) . The clinico-epidemiological features and the results of patient management are shown in Tables 1 and 2 . The origin of patients remains mainly the Songololo Territory , Cataractes District , where the GRH IME/Kimpese is located ( Figure 2 ) . The median age of patients ( 20 years ) was similar for both periods . The proportion of female patients increased significantly from 30% before to 49% after the project was initiated ( p = 0 . 005 ) . In both periods , the majority of BU patients were new cases , yet the proportion of relapse cases amongst all admissions reduced from 32 . 8% to 11 . 6% ( p<0 . 001 ) after 2004 . The proportion of ulcerative forms at admission decreased from 95 . 3% to 85 . 8% after 2004 ( p = 0 . 041 ) , and the proportion of SUF increased from 32 . 8% to 60 . 7% amongst the ulcers ( p<0 . 001 ) ( Figure 3 ) . There was no change in the proportion of confirmed osteomyelitis nor in the proportion of patients presenting with joint movement limitations . The reported median duration of the disease since the appearance of first symptoms increased from 6 to 8 weeks . Globally , the proportion of patients who healed with complications did not change significantly from 23 . 4% to 19 . 5% ( p = 0 . 496 ) , even amongst patients declared cured only , from 31 . 3% to 21 . 0% ( p = 0 . 136 ) . However , the number of cases that healed without complications increased significantly from 51 . 6 to 73 . 2% ( Figure 4 ) ( p = 0 . 001 ) . The proportion of cases confirmed by at least one laboratory test positive for M . ulcerans remained the same ( 70% in 2002–2004 versus 61% in 2005–2007 , p = 0 . 183 ) . Antibiotic therapy was introduced as part of the control project , and was prescribed to 56 . 3% of patients , although most patients continued to receive surgery ( 93 . 7% previously compared to 84 . 2% after 2004 , p = 0 . 052 ) . Ninety patients ( 47 . 4% ) were treated by a combination of antibiotics ( rifampicin and streptomycin ) and surgery . Seventy patients ( 36 . 8% ) were treated with surgery alone , seventeen patients ( 8 . 9% ) only with antibiotics , and thirteen ( 6 . 8% ) were treated with daily wound dressing . The median duration of hospitalization , around 90 days , was approximately similar during both periods ( Table 2 ) and varied by disease category during the second period , respectively 60 days for category I ( Figure 5 and 6 ) , 81 days for category II , and 118 days for category III . The case fatality rate was significantly decreased from 18 . 7% during the previous period ( 12 out of 64 patients ) to 3 . 2% ( 6 out of 190 patients ) during the second period ( p<0 . 001 ) . Conditions associated with mortality among BU patients in the previous period were as follows: sepsis in four patients out of twelve ( 33% ) , malnutrition and anaemia in nine patients ( 75% ) , edematous disseminated disease in two patients ( 16 . 6% ) , postsurgical shock in one patient ( 8% ) , and cancerization in two patients ( 16 . 6% ) .
Overall , the results after 3 years of implementation of BU control activities in Songololo Territory are encouraging . However , the morbidity and disabilities due to BU remain high among our patients . The burden of BU in terms of human suffering , long duration of hospitalization , the development of disabling sequelae , and socio-economic repercussions , is mainly attributable to the late detection of cases . For this reason , secondary prevention through earlier case detection and treatment remains one of the key measures in the control of BU [30] . To reduce the burden and to increase the coverage of the population at risk , we consider that a dedicated BU control program at central and provincial level , that operates in close collaboration with the existing polyvalent health services , would be the most efficient way to organize the control of BU in Songololo Territory . The aforesaid program should involve education of the population in the endemic areas , training of healthcare workers , early detection by active case-finding and adequate case management provided free of charge . Further decentralization and integration of BU control activities may improve access to diagnosis and care at the most peripheral level of the health system . A close collaboration between the BU control project and the health zones is essential for the implementation of a simple , functional , and efficient active surveillance system in a resource-limited context . | Buruli ulcer ( BU ) , which is caused by Mycobacterium ulcerans , is an important disabling skin disease . However , BU has been neglected in many endemic African countries , including in the Democratic Republic of Congo . The province of Bas-Congo contains one of the most important BU foci of the country , i . e . the Songololo Territory in the District of Cataractes . In 2004 a specialized BU control program was launched in that area . The present study aims to evaluate the impact of the above-mentioned program , by comparing clinical profiles and outcomes of the group of patients admitted at the General Reference Hospital ( GRH ) of the “Institut Médical Evangélique” ( IME ) of Kimpese 3 years before the start of the project ( 2002–2004 ) with those admitted during the 3 years after the start of the project ( 2005–2007 ) . The project implementation was associated with a strong increase in the number of admitted BU cases at the GRH and a fundamental change in the profile of those BU patients . Despite these encouraging results , our study provides some limitations of such program , and highlights the need of considering new strategies to better improve BU control in a low resources setting . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [
"medicine",
"infectious",
"diseases",
"public",
"health",
"and",
"epidemiology",
"neglected",
"tropical",
"diseases",
"infectious",
"disease",
"control",
"public",
"health"
] | 2011 | Effect of a Control Project on Clinical Profiles and Outcomes in Buruli Ulcer: A Before/After Study in Bas-Congo, Democratic Republic of Congo |
Chagas disease is a zoonotic parasitic disease well-documented throughout the Americas and transmitted primarily by triatomine ‘kissing bug’ vectors . In acknowledgment of the successful history of vector control programs based on community participation across Latin America , we used a citizen science approach to gain novel insight into the geographic distribution , seasonal activity , and Trypanosoma cruzi infection prevalence of kissing bugs in Texas while empowering the public with information about Chagas disease . We accepted submissions of kissing bugs encountered by the public in Texas and other states from 2013–2014 while providing educational literature about Chagas disease . In the laboratory , kissing bugs were identified to species , dissected , and tested for T . cruzi infection . A total of 1 , 980 triatomines were submitted to the program comprised of at least seven species , of which T . gerstaeckeri and T . sanguisuga were the most abundant ( 85 . 7% of submissions ) . Triatomines were most commonly collected from dog kennels and outdoor patios; Overall , 10 . 5% of triatomines were collected from inside the home . Triatomines were submitted from across Texas , including many counties which were not previously known to harbor kissing bugs . Kissing bugs were captured primarily throughout April-October , and peak activity occurred in June-July . Emails to our dedicated account regarding kissing bugs were more frequent in the summer months ( June-August ) than the rest of the year . We detected T . cruzi in 63 . 3% of tested bugs . Citizen science is an efficient approach for generating data on the distribution , phenology , and infection prevalence of kissing bugs—vectors of the Chagas disease parasite—while educating the public and medical community .
Citizen science—the engagement of non-scientists in collecting scientific data—has long been used in ecological and wildlife research [1–3] , resulting in an engaged public and providing researchers access to data from large geographic and temporal scales . We pose that citizen science is a powerful yet underutilized tool in public health , given that community engagement is recognized as a core component of many successful public health programs [4] . Chagas disease is a vector-borne zoonotic disease caused by the parasitic protozoan Trypanosoma cruzi . Infection with T . cruzi can result in cardiac and digestive disease in humans and dogs that may not manifest until years after infection . Disease in humans is well-documented throughout the Americas [5 , 6] , and canine Chagas disease is well-documented in Texas [7 , 8] . In 2013 and 2014 , the first two years in which Chagas disease was a notifiable disease in Texas , a total of 351 canine cases and 39 human cases were reported; the latter including 12 locally-acquired cases [9 , 10] . Colloquially referred to as ‘kissing bugs’ , triatomine insects ( Fig 1 ) are vectors of Trypanosoma cruzi . In the U . S . , the highest species diversity of triatomines is found in Texas [5 , 11] . Triatomine bugs feed on blood during all stages of their lives , and may acquire the parasite from feeding on an infected mammal . The parasite T . cruzi is spread through the feces of the insect . Community-based vector surveillance has been widespread for decades as an approach to manage Chagas disease in South and Central America , through which householders monitor kissing bug presence within the home to allow for timely response with insecticide treatment . In these regions , some species of kissing bugs occupy a domestic niche ( i . e . , they successfully establish colonies in houses ) [12] . Diverse approaches have been employed in community-based vector surveillance programs , including the use of sensor boxes for passive detection of triatomines [13] and training of community leaders in monitoring for reinfestation and insecticide spraying [14 , 15] . Community-based collections were found to be more sensitive than the gold standard of timed manual searches for triatomine recoveries [16] . A systematic review of Chagas disease vector control interventions across South and Central America concluded that community participatory surveillance significantly boosted vector detection probabilities above those found by vector control program staff using active searches or vector detection devices [17] . Further , retrospective analyses of data from Argentina revealed that vector control strategies that incorporate community participation avert more human cases of disease and cost less than vertical or centralized strategies that consist of insecticide application by program staff only [18] . Community engagement has less commonly been used in the southern U . S . for kissing bug research and control , likely because the vector species in the southern U . S . tend not to colonize homes in the same manner as in Latin America , and insecticide spraying within the home is therefore not a widely used tool for public health protection . The first recruitment of the public in the U . S . to help collect kissing bugs was in 1941 , when Dr . Sherwin F . Wood of Los Angeles City College encouraged Arizona miners to collect insects from their sleeping quarters with the recruitment slogan ‘Nab that bug at one cent each for Dr . Wood at City College to keep . ’ [19]; this was followed by other similar efforts in the 1940s [20 , 21] . Subsequently , community participation in kissing bug surveillance in Tucson , Arizona [22 , 23] and New Orleans , Louisiana [24] has provided unprecedented information on vector phenology and infection in these regions . For public health and research purposes , the recruitment of submissions of kissing bugs from citizens who incidentally encounter them is an attractive option for collection given that kissing bugs are nocturnal , elusive , and difficult to collect using standardized entomological traps [11 , 25 , 26] . Akin to the widespread community-based vector surveillance programs in areas where Chagas disease is endemic across Latin America , here , we describe the implementation and results from a two-year citizen science program in Texas that provides public health education and encourages citizens to aid in the collection of kissing bugs across Texas .
In early 2013 , we developed a citizen science program for Chagas disease research with priority interest in Texas . Our program provides resources for people seeking information about Chagas disease and kissing bugs in the U . S . , while also requesting kissing bug samples through a variety of media: printed pamphlets ( S1 File and S2 File ) , phone communication , an educational website ( http://kissingbug . tamu . edu ) , solicitations on news stations , and a dedicated e-mail address ( kissingbug@cvm . tamu . edu ) . The public may submit insect photos through the website or email to be identified by our team , and are invited to submit kissing bug specimens along with associated information . Submitters are informed about Chagas disease transmission , and cautioned to not touch the insects with bare hands . We request that the bugs be captured in bags and stored in a freezer , to kill the insect before shipping . The minimum requested dataset to accompany each bug includes: date , time and location of capture and whether the bug was alive or dead . Location data were validated and geo-coded in a geographic information system ( ArcMap , ESRI , Redlands , CA ) . In the laboratory , kissing bugs were identified to species [27] , measured , sexed , and dissected . Following DNA extraction ( Omega Bio-tek , Norcross , GA; Qiagen , Germantown , MD ) , bug hindguts were tested for infection through amplification of T . cruzi satellite DNA quantitative real-time PCR that is selective for T . cruzi [28] for which our internal validations defined a positive sample as one with a cycle threshold value of 33 or less . This qPCR is highly sensitive with a limit of detection that approximates 0 . 5 parasite equivalents of DNA [28] . Bugs known to have fed on humans were sent to the Centers for Disease Control via Texas Department of State Health Services for testing so that submitters can be in immediate contact with those who can make medical recommendations . We provided submitters with the species identification and preliminary T . cruzi infection status of their submission , which is shared with a statement about potential false positive or false negative results . Our website includes an interactive map to allow submitters to see their data contributions and examine the spatial and temporal distribution of kissing bugs submitted by the public Texas . We provided citizens with information on reducing kissing bug occurrence in dog kennels or patios outside the home , as these were the primary areas from which bugs were collected . These recommendations included turning off the outdoor lights at night , housing dogs indoors when possible , reducing woody debris or other potential bug harborage sites within the vicinity of the home/kennel , and the use of commercially-available insecticides , although none available in the United States are labeled for the control of triatomines .
From May 2013 through December 2014 , we received approximately 4 , 000 emails that resulted in a total of 1 , 980 kissing bug submissions . The triatomines submitted to the program comprised at least seven species , of which T . gerstaeckeri and T . sanguisuga were most abundant ( 71 . 3 and 14 . 4% of submissions , respectively; Fig 1 and Table 1 ) . Locations from which triatomines were most commonly reported to be collected include dog kennels ( 24 . 6% ) , patios/porches ( 19% ) , and inside homes ( 10 . 8% ) , followed by other locations including outside walls of homes , garages , cat sleeping areas , inside buildings , barns , pools , tents , and chicken coops . Overall , 10 . 8% of triatomines representing 5 species were collected from inside homes ( Table 1 ) and the majority of submissions of adults from inside homes consisted of a single bug that was encountered . As a proportion of the number of collections of each species , T . rubida , T . lecticularia , and T . sanguisuga were most commonly captured inside homes . Regarding nymphs , 20 of the 56 ( 36 . 7% ) nymphs submitted to our program were collected from inside the home ( Table 1 ) , including three submissions of more than one nymph ( two submissions with two nymphs; one submission with five nymphs ) . Over 99% of submissions were from Texas ( 1 , 968 kissing bugs ) , although we also received kissing bugs from Arizona , Florida , Louisiana , Oklahoma , and Virginia ( Fig 2 ) . In our initial program year in 2013 , we received 881 kissing bugs submitted by 145 citizens . Our expanded program in 2014 resulted in the submission of 1 , 099 kissing bugs by 243 citizens , 13 of which had submitted bugs to our laboratory the preceding year . Whereas the majority of citizens submitted a single bug ( 200 individuals ) , many individuals submitted multiple bugs . There were 21 individuals who submitted 20 or more bugs over two years , including one individual who submitted 271 bugs . The majority of these large quantity submitters ( 90 . 5% ) found bugs in the sleeping quarters of their dogs and expressed concerns about canine Chagas disease risk . Kissing bugs received by our laboratory were captured primarily throughout April-October , with the highest number of captures in June-July ( Fig 3 ) . The small number of kissing bugs ( n = 11 ) that were captured in the winter months ( November-March ) were mainly collected from indoors ( 63 . 6% ) and were comprised of a higher percentage of nymphs ( 36 . 4% ) than submissions throughout the summer months ( 26 . 3% ) . In the subset of 694 kissing bugs submitted through this program that we subjected to molecular detection of T . cruzi , we detected infection in 493 bugs ( 63 . 3% ) . In all counties from which at least two bugs were submitted , at least one infected bug was detected . Since establishing a dedicated email account in late 2013 , emails regarding kissing bugs were more frequent in the summer months ( June-August ) than the rest of the year . Periods of exceptionally high email traffic were frequently associated with newscasts and releases of online media related to Chagas disease and kissing bugs ( Fig 4 ) . Captures of non-kissing bugs ( mainly reduvvids and other hemipterans; Fig 5 ) represented approximately 10% of photo submissions to our program . We occasionally received emails from citizens concerned about a bug bite that may be from a triatomine , sometimes accompanied by photos of the bite site . Rarely , these citizens also have collected a kissing bug . In all these cases , we put the citizen in contact with the local contact of the Texas Department of State Health Services who can investigate further and provide medical recommendations when warranted .
We used a citizen science approach to establish a collection of triatomine vectors nearing 2 , 000 specimens in order to define key periods of kissing bug activity , expand the county-level known range of kissing bugs in Texas , and ascertain infection prevalence with T . cruzi . This method of sampling provides unique insight into the specific subset of bugs in nature that are epidemiologically relevant—that is , those bugs that people are encountering during daily activities and that potentially pose the highest risk for transmission of T . cruzi [26] . Further , the educational campaign and community engagement at the core of this initiative allow people to take an active role in understanding how to improve their health . This two-year citizen science program has revealed a similar geographic distribution of kissing bugs in Texas as was previously documented over almost eight decades ( Fig 2 ) . Our results highlight kissing bug occurrence in central and south Texas , which were predicted to be the highest Chagas disease risk zones in a statewide risk map [30] and further extend potential risk zones to include north Texas . Further , the expansive occurrence data from the citizen science initiative can provide unprecedented spatial resolution to complement the limited data used in a previous state-wide effort ( 108 kissing bugs from 63 unique spatial cells ) [30] . We received vector submissions from six of the seven focal areas across Texas from which T . cruzi seropositive dogs were recently detected [8] . The peak in the collection of adult bugs occurred in June-July , with most activity occurring between April-October . While this apparent phenology certainly reflects periods of human outdoor activity given the collection method , it is congruent with the only other study of triatomine phenology in Texas which employed blacklight traps ( a collection method that is independent of public outdoor activity ) in a central Texas county in April-September to conclude that 83 . 4% of adult triatomines collections occurred in May-July [31] . Because vector activity is a key component of human risk , detailed phenology data are useful for public health initiatives . The detected T . cruzi infection prevalence in citizen-submitted Texas kissing bugs was 63% , and is similar to that found in previous studies of kissing bugs from Texas that were collected using other means . For example , a sample of 241 bugs , including those collected from wildlife nests and by health department employees , was characterized by 50 . 7% infection prevalence [11] , and 69–82% of bugs collected from houses and dog kennels in central Texas were infected [25] . The analysis of citizen-collected data presents unique challenges due to observer error and sampling bias [3] . For example , public submission programs result non-target bug species [32 , 33]; however this potential source of observer error is controlled for in our program by laboratory identification of all kissing bugs . The geographic breadth of submissions reflects the area over which citizens are aware of the program and able to contribute to it . While areas from which no bugs were submitted cannot be interpreted as an absence of kissing bugs , the occurrence data are useful for increasing medical and veterinary awareness for Chagas disease over an expanded region . The longitudinal pattern of inquiries from the public revealed that emails to our citizen science account peaked after media events ( Fig 3 ) ; many of these emails concerned inaccurate information on television , internet , or social media . The most common cause of confusion resulted when pictures of common bug species that share some similarity in appearance to kissing bugs , but are not vectors of T . cruzi , were displayed while discussing Chagas disease on the news . Our data demonstrate the influence of the media for increasing awareness for the citizen science initiative to contribute to the growing field of digital epidemiology [34] . This citizen science program has resulted in strengthened relationships among university researchers , state health departments , the Centers for Disease Control and Prevention , clinical veterinarians , medical practitioners , and the general public . Such coordinated efforts among stakeholders—including the public—for insect surveillance offer opportunities for integrated pest management , research , and the protection of human health ( e . g . , The Collaborative Strategy on Bed Bugs [35] and nuisance black fly reporting [36] ) . The collection of samples generated through this program will be available for analyses of triatomine population genetics , blood meal analyses , genetic typing of T . cruzi , and additional research pursuits to enhance our understanding of vector ecology , allowing us to further build upon state-wide and regional models of triatomine distributions and disease risk [30 , 37] . Given the demonstrated public health benefit of community engagement in vector surveillance and control in areas of Chagas disease endemicity across Latin America , citizen science should be promoted as a key approach for enhancing vector-borne disease research and public health protection efforts in the United States . | We created a kissing bug citizen science program in Texas to educate the public about Chagas disease , a vector-borne disease of humans and dogs , and to create a mechanism for the public to submit triatomine ‘kissing bug’ vectors to our research program . From May 2013 to December 2014 , we designed an interactive website , distributed pamphlets , and responded to emails and phone calls from the public . This resulted in the submission of 1 , 980 kissing bugs , mostly collected from dog kennels and outdoor patios , expanding the geographic regions known to harbor kissing bugs in Texas and allowing insight into a cross-section of bugs of high epidemiological and veterinary relevance . Citizen submissions of kissing bugs peaked in June-July and showed 63 . 3% infection prevalence with Trypanosoma cruzi . Citizen science is an efficient mechanism for gaining novel insight into vector occurrence and infection . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] | [] | 2015 | Combining Public Health Education and Disease Ecology Research: Using Citizen Science to Assess Chagas Disease Entomological Risk in Texas |
Human T lymphotropic virus type 1 ( HTLV-1 ) trans-activator/oncoprotein , Tax , impacts a multitude of cellular processes , including I-κB kinase ( IKK ) /NF-κB signaling , DNA damage repair , and mitosis . These activities of Tax have been implicated in the development of adult T-cell leukemia ( ATL ) in HTLV-1-infected individuals , but the underlying mechanisms remain obscure . IKK and its upstream kinase , TGFβ-activated kinase 1 ( TAK1 ) , contain ubiquitin-binding subunits , NEMO and TAB2/3 respectively , which interact with K63-linked polyubiquitin ( K63-pUb ) chains . Recruitment to K63-pUb allows cross auto-phosphorylation and activation of TAK1 to occur , followed by TAK1-catalyzed IKK phosphorylation and activation . Using cytosolic extracts of HeLa and Jurkat T cells supplemented with purified proteins we have identified ubiquitin E3 ligase , ring finger protein 8 ( RNF8 ) , and E2 conjugating enzymes , Ubc13:Uev1A and Ubc13:Uev2 , to be the cellular factors utilized by Tax for TAK1 and IKK activation . In vitro , the combination of Tax and RNF8 greatly stimulated TAK1 , IKK , IκBα and JNK phosphorylation . In vivo , RNF8 over-expression augmented while RNF8 ablation drastically reduced canonical NF-κB activation by Tax . Activation of the non-canonical NF-κB pathway by Tax , however , is unaffected by the loss of RNF8 . Using purified components , we further demonstrated biochemically that Tax greatly stimulated RNF8 and Ubc13:Uev1A/Uev2 to assemble long K63-pUb chains . Finally , co-transfection of Tax with increasing amounts of RNF8 greatly induced K63-pUb assembly in a dose-dependent manner . Thus , Tax targets RNF8 and Ubc13:Uev1A/Uev2 to promote the assembly of K63-pUb chains , which signal the activation of TAK1 and multiple downstream kinases including IKK and JNK . Because of the roles RNF8 and K63-pUb chains play in DNA damage repair and cytokinesis , this mechanism may also explain the genomic instability of HTLV-1-transformed T cells and ATL cells .
Human T-lymphotropic virus type 1 ( HTLV-1 ) is the etiological agent of adult T-cell leukemia/lymphoma ( ATL ) and HTLV-1-associated myelopathy/tropical spastic paraparesis ( HAM/TSP ) . The HTLV-1 genome encodes a trans-activator/oncoprotein known as Tax , which is crucial for viral transcription and cellular transformation [1] . Tax constitutively activates IκB kinases , causing the phosphorylation and degradation of IκBα and activation of the canonical NF-κB pathway . It also up-regulates RelB and p100 ( precursor of NF-κB2 ) expression and increases the phosphorylation and proteolytic processing of p100 to p52 , thus activating the non-canonical NF-κB pathway [2–4] . IKK/NF-κB activation by Tax is causally linked to T-cell transformation and ATL development [5–9] . The mechanisms underlying Tax-induced IKK activation remains unclear , but require the regulatory subunit of IKK: NF-κB essential modulator ( NEMO ) and two IKK kinases: TGF-β-activated kinase 1 ( TAK1 ) and NF-κB-inducing kinase ( NIK ) [10–12] . Polyubiquitin chain assembly is a post-translational modification by which protein-linked or unanchored polyubiquitin chains are synthesized stepwise via reactions requiring three classes of enzymes , E1 activating enzymes , E2 conjugating enzymes and E3 ligases . Lysine 48 ( K48 ) -linked polyubiquitin targets proteins for proteasome-mediated degradation , while lysine 63 ( K63 ) -linked and linear polyubiquitins are crucial for cytokine-mediated IKK/NF-κB activation and DNA damage repair ( DDR ) [13–15] . Upon engagement of receptors by cytokines such as TNFα and IL-1 , members of the TNF receptor-associated factor ( TRAF ) family such as TRAF6 , TRAF2/5 , and cIAP1/2 become recruited to the receptors and function as E3 ligases to assemble free or protein-anchored K63-linked polyubiquitin ( K63-pUb ) chains . The IKK kinase , TGFβ-activated kinase 1 ( TAK1 ) , and IKK both contain ubiquitin-binding subunits ( TAB2/3 and NEMO respectively ) that facilitate their recruitment to K63-pUb chains where TAK1 undergoes auto-phosphorylation/activation . Activated TAK1 then phosphorylates/activates IKK in its vicinity . IKK in turn phosphorylates IκBα , targeting it for K48-linked polyubiquitination and proteasomal degradation , thereby activating the canonical NF-κB pathway . Interestingly , conjugation of Tax by K63-pUb has been reported to correlate with IKK activation , and requires an E2 enzyme known as Ubc13 . Indeed , mouse embryo fibroblasts containing bi-allelic deletion of the Ubc13 gene are deficient in supporting Tax-mediated NF-κB activation [16] . IKK activation by Tax in cytosolic extract was also inhibited by the K63R mutant of ubiquitin [17] . These results support the notion that K63-pUb assembly plays an important role in Tax-mediated IKK/NF-κB activation . The holo Ubc13-containing E2s that carry out K63-pUb chain assembly are heterodimers consisting of a catalytically active Ubc13 and a catalytically inactive E2 variant , Uev1A or Uev2 [18] . The Ubc13:Uev1A heterodimer has been shown to be essential for IKK/NF-κB activation [13] , while the Ubc13:Uev2 heterodimer is thought to be involved in DDR [18 , 19] . Here we show that Ubc13:Uev1A ( or Ubc13:Uev2 ) and the E3 ubiquitin ligase , ring finger protein 8 ( RNF8 ) are crucial for Tax-mediated IKK activation in vivo and in vitro . The ablation or over-expression of RNF8 drastically reduces or augments canonical NF-κB activation by Tax . We demonstrate that Tax directly stimulates RNF8 and Ubc13:Uev1a ( or Ubc13:Uev2 ) to assemble long and free K63-pUb chains in vitro and in vivo . The unanchored K63-pUb chains so assembled then activate TAK1 , which in turn activates IKK and MKKs , and the downstream NF-κB and c-Jun N-terminal kinase ( JNK ) signaling pathways . In the presence of Tax , RNF8 is increasingly redirected from the nucleus to the cytoplasm . Because RNF8 is a key signaling molecule for DNA damage repair and cytokinesis; and because K63-pUb chains play many roles in cell signaling , the inappropriate activation of RNF8 and the over-abundance of K63-pUb chains produced in Tax-expressing cells activate multiple signaling pathways and likely also interfere with DDR and mitosis .
To elucidate the mechanism ( s ) underlying Tax-driven IKK activation , we used a cell-free system [17 , 20] in which the cytosolic extract ( S100 ) of HeLa-G ( Fig 1A ) cells was supplemented with purified hexa-histidine-tagged Tax derived from E . coli . In this system , the addition of Tax and ATP to HeLa-G S100 extract activated IKK , as indicated by the appearance of phospho-IκBα ( p-IκBα ) in immunoblots ( Fig 1A , lane 2 ) . Interestingly , addition of either Ubc13:Uev2 or Ubc13:Uev1A E2 conjugating enzyme complex into the S100 extract further stimulated Tax-dependent IKK activation as evidenced by the increased phosphorylation of IκBα ( lanes 4 & 5 ) , but had no effect in the absence of Tax ( lanes 6–8 ) . Importantly , the same results were observed in S100 extract prepared from Jurkat T cells with quantitative conversion of IκBα to p-IκBα in the presence of Tax ( S2 Fig ) , and addition of other E2 conjugating enzymes , UbcH5b and UbcH5c , had no effect on Tax-mediated IKK activation ( S2 Fig ) . When S100 extract was prepared from HeLa-G cells depleted of Ubc13 , Uev2 , or Uev1A by shRNA-mediated knockdown , IKK activation by Tax in vitro was significantly diminished ( Fig 1B p-IκBα blot in lane 2 vs lanes 4 , 6 , & 8 ) . The addition of purified Ubc13 , Uev2 or Uev1A to the respective depleted S100 lysates restored IKK activation by Tax ( Fig 1C lanes 3 vs 2; 6 vs 5; 9 vs 8 ) . The extent of NF-κB activation by Tax as measured by the E-selectin-Luc reporter assay was also reduced in HeLa-G cells deficient in Ubc13 , Uev2 , or Uev1A ( Fig 1D ) . The moderate impact of these knockdowns is likely due to NF-κB activation contributed by residual Ubc13 E2 complexes as well as the non-canonical NF-κB pathway ( see below ) . These results agree with previous studies showing that Ubc13 is critical for Tax-mediated IKK/NF-κB activation [16] . Tax is not an E3 ligase however . No ubiquitin chain assembly could be detected in reaction mixtures containing only Tax , Ubc13:Uev1A ( E2 ) , E1 , Ub , and ATP ( see below ) . This prompted us to consider the possibility that a cellular E3 ligase may be recruited by Tax for IKK activation . A search in the literature for E3 ligases that specifically utilize Ubc13 for ubiquitin chain assembly found C-terminus of HSC70-interacting protein ( CHIP ) [21] , checkpoint with forkhead and RING finger domains ( CHFR ) [22] , helicase-like transcription factor ( HLTF ) [23] , RING finger protein 8 ( RNF8 ) [24] , TNF receptor-associated factor 2 ( TRAF2 ) , TRAF5 , and TRAF6 ( reviewed in [25] ) to be of interest . Of these E3 ligases , RNF8 caught our attention . RNF8 is a E3 ligase that contains an Forkhead-Associated ( FHA ) domain for binding specific phospho-proteins , a coiled-coil region responsible for dimerization , and a RING domain for E2 binding and ubiquitin chain assembly ( Fig 2C ) [26–28] . RNF8 is important for DDR [29 , 30] , centrosomal functions , and cytokinesis [31 , 32] , all of which are known to be disrupted by Tax [33] . Furthermore , Tax is a dimer that interacts with and stabilizes the coiled-coil domains of dimeric bZip transcription factors CREB and ATF1 [34 , 35] . For these reasons , we tested RNF8 as a potential target of Tax . Indeed , RNF8 co-purifies with IKKα , IKKβ , TAK1 , and Tax when S-peptide-tagged ( S-tagged ) -Tax is captured from transfected 293 cell lysate using RNase S-agarose beads ( Fig 2A ) . Notably , a fraction of Tax was modified ( Fig 2A left panel ) with size increments consistent with polyubiquitination reported previously [16] . To investigate the role of RNF8 in Tax-induced IKK activation , glutathione S-transferase ( GST ) -RNF8 and GST-RNF8345–485 fusion proteins were purified and treated with 3C protease to remove GST , yielding highly purified full-length and truncated RNF8 ( Fig 2B and 2C ) . As an indication that RNF8 is critical for IKK activation by Tax , the S100 extract prepared from HeLa-G cells depleted of RNF8 by shRNA-mediated knockdown ( RNF8KD ) was unable to support IKK activation by Tax ( Fig 2D p-IκBα blot lane 2 vs 4 ) . Notably , while the addition of exogenous RNF8 to the HeLa-G S100 extract in the absence of Tax had no effect on IKK ( Fig 2E lane 4 ) , exogenous RNF8 dramatically enhanced Tax-mediated activation of IKK ( Fig 2E lane 3 vs 2 ) . As expected , RNF8 addition to S100 extract prepared from RNF8KD cells restored Tax-mediated IKK activation ( Fig 2E lane 6 vs 7 ) while the truncated RNF8345–485 failed to do so ( Fig 2E lane 9 vs 7 ) , suggesting that the coiled-coil region and RING domain of RNF8 are not sufficient to effectively enhance Tax-mediated IKK activation . To confirm the crucial role of RNF8 in Tax-mediated IKK activation , an RNF8-null HeLa cell line ( RNF8KO ) was generated via the CRISPR/Cas9 system ( Fig 3A ) . As anticipated , the loss of RNF8 reduced NF-κB activation by Tax to ∼1/3 to ∼1/2 of that in wild-type HeLa-G cells ( Fig 3B shaded vs open bars ) . Co-transfection of Tax with increasing amounts of RNF8 in HeLa-G cells also augmented the already potent NF-κB activation by Tax in a dose-dependent manner ( Fig 3C open bars ) . Finally , transfection of RNF8KO cells with RNF8 DNA rescued Tax-induced NF-κB activation ( Fig 3C shaded bars ) . Interestingly , in RNF8KO cells , Tax continued to promote p100 phosphorylation and processing ( Fig 3D p-p100 IB lanes 2 & 5 vs 1 & 4 and 3 & 6 ) , suggesting that the non-canonical pathway is activated by Tax via an RNF8-independent mechanism . We note that the level of Rnf8 is reduced in Ad-Tax-transduced cells . This is likely due to the cell cycle arrest/senescence caused by Tax . In RNF8KO cells , the extent of Tax-driven IκBα degradation was reduced but not abrogated ( Fig 3D IκBα IB lanes 2 & 5 vs 1 & 4 and 3 & 6 ) . Whether this is caused by IKKα and/or the K63-pUb conjugated to Tax ( presumably by a different E3 ligase ) remains to be determined . In the S100 extract deficient in TAK1 ( TAK1KD ) , IKK activation by Tax and RNF8 is greatly diminished ( Fig 3E lanes 6 & 7 vs 2 & 3 , p-IKKα/β blot ) . By contrast , in wild-type HeLa-G S100 extract , p-IκBα , p-IKKα/β , and p-TAK1 levels were dramatically enhanced by Tax and RNF8 in combination ( Fig 3E lane 3 ) , supporting the notion that Tax activates the canonical NF-κB pathway via RNF8 , TAK1 , and IKK . Most interestingly , significant p-JNK but not p-p38 kinase was readily detected in reactions containing exogenous Tax and RNF8 ( Fig 3E lane 3 ) , reminiscent of the constitutive activation of JNK in HTLV-1- or Tax-transformed cells reported previously [36] . We think this is likely due to Tax/RNF8-activated TAK1 effecting activation of MKK7 and its downstream target , JNK [37] . In aggregate , these data strongly suggest that the TAK1 activated by Tax/RNF8/Ubc13:Uev1A/Uev2 concurrently activated multiple downstream kinases including IKK and kinases upstream of JNK , possibly MKK7 . For reasons unclear at present , a deficiency in TAK1 caused moderate p38 kinase and JNK activation as suggested by the presence of p-p54 and p-p46 SAPK/JNK and p-p38 in the S100 extract irrespective of Tax ( Fig 3E bottom panel lanes 5–8 ) . RNF8 is an E3 ligase that localizes primarily to the nucleus during interphase where it mediates DDR [30 , 38] . Since activation of TAK1 and IKK occurs in the cytoplasm , and Tax is known to shuttle between nuclear and cytoplasmic compartments [39] , we asked if Tax might affect the cellular localization of RNF8 . The Tax reporter cell line , HeLa-G [40] , was transduced with Ad-Tax for 48 hours and the sub-cellular distribution of RNF8 determined by immunofluorescence . As indicated in Fig 4A , strong nuclear RNF8 signal ( Red ) could be readily detected in untransduced HeLa-G cells ( GFP- ) . In contrast , in Tax-transduced cells ( Fig 4A , upper right panel , GFP+ ) , cytoplasmic RNF8 signal was increased ( upper left panel ) . Subcellular fractionation of Tax-transduced HeLa-G cells also showed more RNF8 in the cytosolic than the nuclear fraction ( Fig 4B lane 3 vs 4 ) . HeLa-G cells transduced with the control Ad-tTa vector , by contrast , had an even distribution of RNF8 in both cytosolic and nuclear fractions ( Fig 4B lane 1 vs 2 ) . Further examination of HTLV-1-unrelated Jurkat and HTLV-1-transformed MT4 T cell lines by immunofluorescence ( Fig 4C ) and subcellular fractionation ( Fig 4D ) indicate that a significant fraction of RNF8 localizes to the cytosol of T cells , and Tax expression correlates with increased cytoplasmic distribution of RNF8 . These results support the notion that Tax can redirect RNF8 to the cytoplasm for the activation of the canonical NF-κB pathway . Because K63-pUb plays a key role in IKK activation , we set out to determine if Tax directly impacts on K63-pUb chain assembly by RNF8 and Ubc13:Uev1a/Uev2 . To this end , RNF8 and RNF8345–485 were incubated with E1 , ubiquitin , ATP , Ubc13:Uev1A or Ubc13:Uev2 , in the presence or absence of Tax in vitro . In the absence of Tax , a low level of polyubiquitin chain assembly by RNF8 and Ubc13:Uev1A or Ubc13:Uev2 could be detected in vitro ( Fig 5A upper panel , lane 3 vs 2 for Ubc13:Uev1A; lane 8 vs 7 for Ubc13:Uev2 ) , and Ubc13:Uev1a was more efficient than Ubc13:Uev2 in supporting polyubiquitin chain assembly ( Fig 5A upper panel , lane 3 vs 8 ) . Remarkably , the addition of Tax greatly stimulated polyubiquitin chain formation by RNF8 and Ubc13:Uev1A or Ubc13:Uev2 ( Fig 5A upper panel , lane 4 vs 3; lane 9 vs 8 ) . Tax also activated polyubiquitin chain assembly by RNF8345–485 ( Fig 5A upper panel , lane 6 vs 5 , for reactions containing Ubc13:Uev1A; lane 11 vs 10 , Ubc13:Uev2 ) , albeit the E3 ligase activity of RNF8345–485 and the extent of its activation by Tax were substantially lower compared to when full-length RNF8 was used ( Fig 5A upper panel , for reactions with Ubc13:Uev1A: lane 6 vs 4 [with Tax] , lane 5 vs 3 [without Tax] , with Ubc13:Uev2: lane 11 vs 9 [with Tax] , lane 10 vs 8 [without Tax] ) . The polyubiquitin chains assembled by RNF8 in the presence of Tax are of greater lengths compared to when RNF8345–485 was used , as revealed by analyzing the samples from the upper panel in a 4–20% polyacrylamide gradient gel ( Fig 5A middle panel , lane 4 vs 6 ) . As expected , RNF8345–485 failed to support IKK activation by Tax in the HeLa-G S100 extract ( Fig 2E lane 9 vs 10 ) , suggesting that the assembly of long K63-pUb chains is crucial for IKK activation . We also found Ubc13:Uev1A to be more robust than Ubc13:Uev2 for RNF8-driven long polyubiquitin chain assembly in vitro ( Fig 5A middle panel , lane 4 vs 9 ) , consistent with the reported activities of Ubc13:Uev1A and Ubc13:Uev2 [18] . Since a fraction of Tax is covalently modified by K63-pUb in vivo , we examined if Tax became polyubiquitinated by RNF8 and Ubc13 in vitro . The polyubiquitination reaction was carried out over a course of 8 hours . Even though Tax and RNF8 greatly stimulated IKK activity within 1 hour after their addition to the S100 extract , and Tax-driven K63-pUb chain assembly ( by Ubc13:Uev1A and RNF8 ) in vitro peaked 4 hours into the in vitro reactions , only a hint of slower-migrating forms of Tax could be detected at 8 hours into the in vitro ubiquitin assembly reaction ( S3 Fig ) . Thus the ubiquitination of Tax by RNF8 and Ubc13 in vitro does not correlate with IKK activation and is likely non-physiological . The polyubiquitination of Tax in vivo is likely carried out by an E3 ligase distinct from RNF8 . Importantly , none of the other proteins ( Uev1A , Uev2 , and RNF8 ) in the in vitro reactions became polyubiquinated ( S4 Fig ) . Thus the K63-pUb chain assembled by Tax/RNF8/Ubc13:Uev1A is unanchored and directly activates TAK1 , and then IKK . These data agree with previous findings showing that free K63-pUb chains assembled by TRAF6 and Ubc13:Uev1A can activate TAK1 [41] . In agreement with the specificity of Ubc13 E2 enzyme complexes , the polyubiquitin chains assembled via RNF8 in the presence of Tax reacted to a K63-pUb-specific antibody ( Fig 5A lower panel , lanes 3 , 4 , 6 , 8 , 9 ) . Furthermore , only wild-type ubiquitin and ubiquitin mutants that contain lysine at amino acid residue 63 such as K48R and K63-only ( all lysine residues except K63 mutated to arginine residues ) ( Fig 5B lanes 2 , 4 , and 5 ) , but not those with altered K63 residue such as K63R , K48-only , and K0 ( all lysine residues substituted with arginine residues ) mutants ( Fig 5B lanes 3 , 6 and 7 ) supported Tax-induced polyubiquitin assembly in vitro . Finally , co-transfection of Tax with increasing amounts of RNF8 into HeLa cells stimulated in a dose-dependent manner the assembly of polyubiquitin chains that reacted with a K63-pUb-specific antibody ( Fig 5C lanes 2–5 vs 1 & 7 ) , while RNF8 alone had no effect ( Fig 5C lane 6 ) . Altogether , these results indicate that Tax usurps RNF8 and Ubc13:Uev1A/2 to assemble K63-pUb chains for the activation of TAK1 , IKK , and other signaling pathways ( summarized in Fig 6 ) .
Many cancer viruses encode regulatory proteins ( Epstein-Barr Virus LMP1 , Kaposi Sarcoma Herpesvirus vFLIP , and HTLV-1 Tax ) that activate IKK/NF-κB as a part of their oncogenic program , but the underlying molecular mechanisms remain incompletely understood . This hampers a clear understanding of the oncogenic processes and the development of treatment and therapeutic approaches . In this study , we have demonstrated in vitro and in vivo that Tax hijacks RNF8—a ubiquitin E3 ligase best known for ubiquitinating histones to signal DDR—to assemble K63-pUb chains for TAK1 , IKK , and JNK activation ( Fig 6 ) . Depletion of RNF8 or constituents of the E2 conjugating enzymes , including Ubc13 , Uev1A and Uev2 , that RNF8 utilizes for K63-pUb assembly diminished while exogenous addition of RNF8 and Ubc13 greatly augmented Tax-induced IκBα phosphorylation in a cell-free system . These in vitro results were correlated with cell-based reporter assays . Importantly , using an in vitro system reconstituted with purified components , we have found that Tax dramatically activated RNF8 and Ubc13:Uev1a/Uev2 to assemble long and unanchored K63-pUb chains . Over-expression of RNF8 in the presence of Tax in vivo also led to a dose-dependent increase in K63-pUb synthesis . Activation of the non-canonical NF-κB pathway by Tax , however , is RNF8-independent and continues to occur in RNF8-null cells . These results demonstrate that Tax usurps cellular ubiquitination machinery to assemble K63-pUb chains for TAK1 , IKK , and canonical NF-κB activation . Interestingly , Tax and RNF8 in combination also activated JNK phosphorylation in vitro . This is consistent with the role of TAK1 in regulating NF-κB-unrelated signaling pathways and provides an explanation for the pleiotropic effect of Tax . Over the past few years , the theme of K63-pUb chains serving as signaling scaffolds has emerged ( reviewed in [42–44] ) . Upon TNF-α or IL-1 stimulation , E3 ubiquitin ligases such as TRAF6 , TRAF2/5 , and cIAP1/2; and signaling molecules such as IRAK1 and RIP1 are recruited to the activated receptors where extensive K63 polyubiquitination takes place on TRAFs , RIP1 , IRAK1 and other molecules [42–44] . K63-pUb chains then serve as signaling platforms for multiple kinases to convene and interact . A recent study has also shown that during IL-1β signaling , linear polyubiquitin ( M1-pUb ) assembled by a unique E3 ligase complex known as LUBAC ( linear ubiquitin assembly complex ) that consists of HOIP , HOIL , and Sharpin becomes covalently attached to K63-pUb to form K63/M1-pUb hybrid [45] . As NEMO , the regulatory subunit of IKK , has 100-fold higher binding affinity for M1-pUb than for K63-pUb , and TAB2/3 , subunits of the holo-TAK1 complex , bind K63-pUb specifically , it is proposed that TAK1 and IKK respectively are recruited to K63-pUb and M1-pUb of the hybrid pUb such that TAK1 becomes auto-activated and signaling between TAK1 and IKK can occur [45] . Whether the assembly of K63-pUb triggers obligatory M1-pUb chain formation and whether M1-pUb chains are needed for Tax-mediated canonical NF-κB activation remain to be determined . Both protein-linked and free unanchored K63-pUb chains have been reported to activate TAK1 and IKK complexes [41] . In the reconstituted system described herein , the K63-pUb chains assembled are free and unconjugated to protein factors used in the assay including Tax , RNF8 and Ubc13 . Tax has been shown to be modified by ubiquitination and sumoylation [46 , 47] . Since Tax is not polyubiquitinated by RNF8 in vitro ( S3 Fig ) , its polyubiquitination may require other E3 ligases . Other cellular factors including RNF4 and NRP/Optineurin have been reported to play a role in Tax-mediated NF-κB activation [48 , 49] . We have found that under the condition of our experiment , RNF4 moderated NF-κB activation by Tax ( S5 Fig ) . This may be related to the K48-linked polyubiquitination and degradation of Tax promoted by RNF4 [50] . Whether and how NRP/Optineurin impacts the Tax-RNF8 signaling axis remains to be determined . RNF8 is involved in the early signaling events of the DNA double-stranded break repair pathway . Via its N-terminal FHA domain , RNF8 is targeted to the ataxia telangiecstasia mutated ( ATM ) -phosphorylated mediator of DNA damage checkpoint 1 ( MDC1 ) that binds to phospho-histone H2 variant H2ax ( γ-H2ax ) accumulating at the site of DNA double-strand breaks where RNF8 promotes K63 polyubiquitination of histones H2a , H2b and γ-H2ax . This then leads to the recruitment of another E3 ligase , RNF168 . The extensive K63-polyubiqutination of histones by RNF8 and RNF168 facilitates the recruitment of the p53-binding protein 1 ( 53BP1 ) and the breast cancer susceptibility protein ( BRCA1 ) for DDR [24 , 29 , 30 , 51] . RNF8 has also been shown recently to localize to sites of cell division where it stimulate K63 polyubiquitination of septin 7 for cytokinesis [31] . Tax-RNF8 interaction therefore may play a role in DDR [52] and cytokinesis [53] defects induced by Tax . Since RNF8 ablation did not cause overt cytological abnormalities in HeLa-G cells , it appears that the cytopathic effects of Tax cannot be explained based solely on RNF8 sequestration . Whether the over-production of mislocalized K63-pUb chains as stimulated by Tax may sequester cellular factors crucial for DDR and cytokinesis is currently under investigation . As the regulation of RNF8 and related RING-domain E3 ligases is poorly understood , elucidating how Tax interacts with and activates RNF8 will provide critical insight into how this class of E3 ligases is regulated . Finally , a clear understanding of how Tax impacts cellular signaling will shed light on the development of ATL and facilitate the design of therapeutic approaches .
Knockdown of each of the Ubc13 , Uev2 , Uev1A and RNF8 genes in a Tax reporter HeLa/18x21-EGFP ( HeLa-G , derived in the lab ) cell line was performed as previously described [54 , 55] . The sequences targeted for each gene are listed in S1 Table ) . Stable cell clones with knockdown of each gene were isolated by limiting dilution and gene silencing validated by immunoblotting . RNF8 knock out cell line was generated using the CRISPR/Cas9 system as described in [56] . Briefly , two complementary oligonucleotides , CACCGTCACAGGAGACCGCGCCGG and AAACCCGGCGCGGTCTCCTGTGAC , corresponding to the 5’ coding region of human RNF8 were synthesized . After annealing , the double-stranded DNA fragment was cloned into the CRISPR/Cas9 vector pX330 ( Addgene ) that had been cut with BbsI . The recombinant plasmid is transfected into HeLa cells by electroporation . Stable clones with RNF8 knockout were screened by immunoblotting after limiting dilution . HEK293T ( ATCC ) , HeLa-G , and HeLa-G knockdown and knockout cell lines were cultured in DMEM supplemented with 10% fetal bovine serum , L-glutamine , 100U/ml penicillin and streptomycin and maintained in 5% CO2 at 37°C . Jurkat T cells ( ATCC ) were grown in RPMI with the same supplements . Cells were harvested and lysed in lysis buffer ( Cell Signaling ) . Routinely , a total of 10–20 μg of proteins is loaded per sample . HTLV-1 Tax hybridoma monoclonal antibody 4C5 was as previously described [53] . Ubc13 , IKKα , IKKβ , TAK1 , p100/p52 , p-IκBα , p-IKKα/β , p-TAK1 , p-JNK/SAPK , p-p38 , p-p100 antibodies were from Cell Signaling Technology . IκBα , RNF8 , HDAC1 , GAPDH , β-actin , HA and ubiquitin antibodies were from Santa Cruz Biotechnology . Uev2 antibody was from Abcam . K63 ubiquitin antibody was from eBioscience . Poly-Histidine antibody was from Sigma-Aldrich . HEK293T cells were transfected with a PiggyBac transposon-based plasmid ( a kind gift from Dr . Pentao Liu [57] ) for Tax-S-Tag . Cells were harvested 48 hours later and lysed with lysis buffer containing a deubiquitinase inhibitor , PR619 ( Life Sensors ) . Cleared cell lysate was then incubated with S-protein agarose beads ( Novagen ) at 4°C overnight . The beads were washed three times with lysis buffer and the protein was then eluted in equal volume of 2X Laemmli Sample Buffer ( Sigma-Aldrich ) . The eluted protein was heated to 95°C for 10 minutes for immunoblotting . HeLa-G cells were co-transfected with E-Selectin luciferase NF-κB reporter plasmid and BC12-Tax , pcDNA-RNF8 and/or RNF4-mCherry by lipofection ( using Promega FuGENE transfection reagent ) for 48 hours . DNA transfection was typically carried out in triplicate in a 24-well plate seeded the night before with 5x104 HeLa-G cells per well in 0 . 5 ml DMEM supplemented with 10% fetal bovine serum . Each transfection contained E-Selectin luciferase ( 250 ng/ml ) , BC-12 Tax ( 50 ng/ml , unless otherwise indicated ) . The final DNA amount per well is adjusted to 1 μg/ml using pcDNA plasmid . The Luciferase Reporter Assay was performed following manufacturer’s protocol ( Promega ) . The luminescent signals were detected using Glomax Multi Detection System ( Promega ) . To detect K63-pUb formation as a function of Tax and RNF8 expression , 1x105 cells/well were grown in 1 ml medium in a 12-well plate and transfected with the indicated amounts of DNA . S100 was prepared as in [20] . Briefly , 1 . 5 x 108 cells were re-suspended in 500 μL of a hypotonic buffer and homogenized using a Dounce homogenizer . Cleared supernatant ( S100 ) was collected after ultracentrifugation at 100 , 000 g for 1 hour . Cell-free assay was performed as in [17] . The S100 cytosolic extract ( ~20 μl at a protein concentration of at least 10 mg/ml ) was incubated with 0 . 5 μM recombinant TaxH6 , 0 . 35 μM recombinant RNF8 , 0 . 5 μM Ubc13/Uev2 ( Life Sensors ) and/or 0 . 5 μM Ubc13H6/Uev1A ( Boston Biochem ) as indicated in an ATP-containing buffer at 30°C for 1 hour . The reactions were quenched by adding 2X Laemmli Sample Buffer ( Sigma Aldrich ) and heated to 95°C for 10 minutes for analysis by immunoblotting . Ubiquitination assay was carried out as in [27] . A ubiquitination reaction typically contains 25 μM wild-type or mutant ubiquitin ( Life Sensors ) , 0 . 1 μM human E1 ( Life Sensors ) , 0 . 4 μM His6-Ubc13/Uev1A ( Boston Biochem ) or Ubc13/Uev2 ( Life Sensors ) , 0 . 75 μM recombinant E3 ligase ( RNF8 or RNF8345-485 ) and/or 0 . 5 μM recombinant Tax in 20–25 μL of ubiquitination buffer ( 20 mM HEPES pH 6 . 8 , 200 mM NaCl , 2 . 5 mM MgSO4 , 10 μM ZnSO4 , 0 . 1 mM DTT , 2 mM ATP ) and incubated at 37°C for 4 hours or the indicated times . The reactions were then quenched and immunoblotted as above . HeLa-G cells were plated on chamber slides and transduced with Ad-Tax ( MOI of 0 . 5 ) for 48 hours . T cells were plated on poly-L-lysine-coated cover slips for 10 minutes . They were then fixed with 4% paraformaldehyde and permeabilized with 0 . 2% Triton X-100 . Cells were immunostained overnight with RNF8 primary antibody ( Santa Cruz Biotechnology ) followed by Alexa Fluor 568 secondary antibody ( Invitrogen ) . The slides were then mounted in Dako Fluorescence Mounting Medium ( Agilent Technologies ) and set at room temperature for 1 hour in the dark . Images were captured using an Olympus IX81F fluorescence microscope or a Pascal confocal microscope . HeLa-G cells were transduced with Ad-Tax or Ad-tTa ( MOI of 20 ) for 48 hours . Cells were harvested and immediately fractionated using a Nuclear and Cytoplasmic Extraction kit by Thermo Scientific . The fractions were then immunoblotted for the indicated proteins . Hexahistidine-tagged Tax protein ( TaxH6 ) was expressed and purified as previously described [58] . Recombinant GST-tagged human RNF8345-485 [27] was expressed in E . coli BL21 DE3 after IPTG induction . Cell pellet obtained from 2 liters of culture was re-suspended in lysis buffer ( 50 mM Tris pH 8 . 0 , 150 mM NaCl , 1 mM DTT and protease inhibitor cocktail ) and lysed using a French Press . The lysate was cleared by centrifugation at 43 , 000 rcf for 30 minutes . Cleared lysate was incubated with 200–300 μL 50% glutathione agarose bead slurry at 4°C overnight and the beads were washed with lysis buffer . GST-RNF8345-485 was then cleaved on the beads with PreScission Protease ( GE HealthCare ) to release RNF8345-485 . Recombinant GST-tagged full-length human RNF8 ( GST-RNF8 ) was expressed in Sf9 cells ( from G . Dveksler ) after infection with a baculovirus vector ( a kind gift from Dr . Titia Sixma [28] ) . Cell pellet was re-suspended in lysis buffer ( 30 mM HEPES pH 8 . 0 , 250 mM NaCl , 10% glycerol , 1 μM ZnCl2 , 1 mM TCEP , and protease cocktail ) and lysed over a salt-ice bath by sonication ( microtip , 60% duty cycle , 15 seconds on and 30 seconds off for 5 minutes ) . The lysate was cleared by centrifugation at 43 , 000 rcf for 30 minutes . Cleared lysate was incubated with glutathione agarose beads at 4°C for 2 hours and the beads were washed with lysis buffer . GST-RNF8 was eluted with 10 mM reduced glutathione and cleaved with PreScission Protease . The GST moiety was removed by passing the reaction mixture through glutathione agarose beads . Recombinant hexa-histidine-tagged Tax ( TaxHis6 ) was expressed in E . coli BL21 DE3 and purified using a HisTrap nickel column ( GE Healthcare ) with an FPLC as previously described [58] except cell lysis was carried out using a French press . The purified TaxHis6 was dialyzed ( 20 mM Hepes pH 7 . 9 , 100 mM KCl , 0 . 5 mM DTT , 0 . 2 mM EDTA , 0 . 5 mM PMSF , 20% glycerol ) and stored frozen at -80°C . Total mRNA was isolated from HeLa-G and Uev1A knockdown cell clones using TRIzol Reagent ( Ambion ) according to manufacturer's instructions . Turbo DNA-free kit ( Ambion ) was used to remove contaminating genomic DNA . Complementary DNA ( cDNA ) was then synthesized using iScript reverse transcription super mix ( Biorad ) . Real-time PCR was performed using the cDNA as templates with Uev1A specific primers purchased from BioRad or β-actin specific primers ( S2 Table ) , and LightCycler DNA SYBR Green I master mix ( Roche Applied Science ) in a LightCycler thermal cycler ( Roche Diagnostics ) . The mRNA level in each sample was normalized to that of the β-actin mRNA . Relative mRNA levels were calculated using the 2−ΔCt method [59] . | Activation of the NF-κB family of transcription factors by the HTLV-1 oncoprotein , Tax , is causally linked to adult T cell leukemia ( ATL ) development in HTLV-1-infected individuals , but the underlying mechanisms are not fully understood . NF-κB activation requires the phosphorylation of its inhibitor , IκBα , by IκB kinase ( IKK ) , which marks IκBα for degradation . In this study , we demonstrate that Tax inappropriately activates a ubiquitin E3 ligase , RNF8 , and ubiquitin E2 conjugating enzymes , Ubc13:Uev1A/Uev2 , to assemble long lysine 63-linked polyubiquitin ( K63-pUb ) chains , which function as signaling platforms for polyubiquitin-binding TGFβ-activated kinase 1 ( TAK1 ) and IKK to congregate and become activated . Because TAK1 mediates the activation of multiple downstream signaling pathways , the mechanism described here can explain the complex effect of Tax on cell signaling . The major functions of RNF8 are to signal cellular DNA damage repair ( DDR ) and cell division by assembling K63-pUb chains at the site of DNA damage and cell cleavage . As such , the inappropriate activation of RNF8 and the over-abundance of K63-pUb chains in Tax-expressing cells may explain how Tax causes DNA damage and cell division defect . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [] | 2015 | HTLV-1 Tax Stimulates Ubiquitin E3 Ligase, Ring Finger Protein 8, to Assemble Lysine 63-Linked Polyubiquitin Chains for TAK1 and IKK Activation |
Recent studies are starting to show that genetic control over stochastic variation is a key evolutionary solution of single celled organisms in the face of unpredictable environments . This has been expanded to show that genetic variation can alter stochastic variation in transcriptional processes within multi-cellular eukaryotes . However , little is known about how genetic diversity can control stochastic variation within more non-cell autonomous phenotypes . Using an Arabidopsis reciprocal RIL population , we showed that there is significant genetic diversity influencing stochastic variation in the plant metabolome , defense chemistry , and growth . This genetic diversity included loci specific for the stochastic variation of each phenotypic class that did not affect the other phenotypic classes or the average phenotype . This suggests that the organism's networks are established so that noise can exist in one phenotypic level like metabolism and not permeate up or down to different phenotypic levels . Further , the genomic variation within the plastid and mitochondria also had significant effects on the stochastic variation of all phenotypic classes . The genetic influence over stochastic variation within the metabolome was highly metabolite specific , with neighboring metabolites in the same metabolic pathway frequently showing different levels of noise . As expected from bet-hedging theory , there was more genetic diversity and a wider range of stochastic variation for defense chemistry than found for primary metabolism . Thus , it is possible to begin dissecting the stochastic variation of whole organismal phenotypes in multi-cellular organisms . Further , there are loci that modulate stochastic variation at different phenotypic levels . Finding the identity of these genes will be key to developing complete models linking genotype to phenotype .
The link between genotype and phenotype is often considered to be deterministic such that a single genotype functions to yield a specific phenotypic value . This deterministic relationship is a central tenet of the desire to develop predictive models allowing an organism's phenotype to be forecasted upon knowing its specific genotype . This deterministic hypothesis is supported by research showing that cells limit stochastic noise/variance in genetic , metabolic , and signaling networks through network topology , a characteristic that is known as network robustness [1]-[6] . This robustness is an inherent property of genetic networks . In evolutionary theory , robustness is predominantly described as canalization wherein genes function to minimize the variance ( maximize the robustness ) of a phenotype [7]–[11] . A well-studied example of genetic control over variance for diverse phenotypes is the heat-shock protein 90 which plays a major role in canalizing existing natural variation [12]–[14] . While a deterministic link between genotype and phenotype is the most frequently studied aspect of evolution and genetics , there is growing research showing the potential evolutionary benefit of a stochastic link between genotype and phenotype . A stochastic link between phenotype and genotype allows an individual genotype to generate a range of phenotypes within a specific environment and causes the portfolio effect wherein the fitness of a specific genotype is determined by the range of phenotypes that it can obtain [15] . In some bacterial settings , stochastic switching of the genotype-to-phenotype link is the evolutionary optimal response to rapid unpredictable environmental fluctuations [16]–[20] . Similarly in single-celled and multicellular eukaryotes , there is beginning to be studies finding polygenic natural variation that determines stochastic noise of gene expression [21]–[25] . This includes Arabidopsis thaliana loci that are known to be under natural selection suggesting that the stochastic aspects of these loci may impart an evolutionary benefit [24] , [26] , [27] . One possible evolutionary benefit of this phenomenon to higher-eukaryotes is that stochastic noise in defense phenotypes can delay the evolution of counter-resistance in biotic pests [28] , [29] . Thus , there is just beginning to be an appreciation of genetic variation controlling stochastic noise in eukaryotic gene expression , which may play a beneficial role in the evolution of these organisms [12]–[14] , [20] [21]–[25] . Similar to transcriptional networks , metabolic networks are thought to be highly structured to maximize deterministic relationships and minimize stochastic variance that could disconnect pathways and potentially generate toxic intermediates [30] . Metabolic robustness is thought to arise from the fact that metabolism is highly interconnected with numerous feedback loops and parallel pathways involving enzymes encoded by both the nuclear and organellar genomes in eukaryotes [31] . This hypothesis was supported by a recent modelling approach where only a few enzymes were predicted to influence stochastic variation in the whole metabolic network [32] . In contrast , a different modelling effort found that stochastic noise can arise in local areas of a metabolic network without spreading throughout the system . This suggests that stochastic variation in the metabolome could be caused by numerous independent loci . [33] However , a lack of empirical evidence on the level or presence of genetically-controlled stochastic variation within metabolism prevents a direct comparison of these two models [24] . To empirically measure the potential for genetic variation to control stochastic variation within the metabolomic network , we measured metabolome variation in a recombinant inbred line ( RIL ) population of Arabidopsis thaliana . Arabidopsis is a key organism in the study of complex traits including the genetic programming of stochastic variation through the use of systems biology and quantitative genomics approaches [24] , [34]–[41] . Additionally , Arabidopsis has been a model system to study the quantitative basis of metabolomic variation in a number of structured and unstructured populations [42]–[44] . Combined with extensive whole genome sequence of natural accessions , this provides the ability to rapidly develop and test hypotheses , as well as find causal genes underlying specific loci of interest [45]–[49] . Finally , there are a large number of existing homozygous populations to enable this analysis [50] . This makes Arabidopsis an ideal system to search for the genetic and molecular basis of complex phenotypes , such as stochastic noise , in higher organisms . Using a replicated , randomized sampling design , we measured metabolome variation in the Kas x Tsu RIL population and compared the quantitative genetics for average metabolite accumulation versus the stochastic variation [24] , [51] . The independently replicated analysis of CV allows us to separate stochastic variance from non-additive variance affecting the mean . This is in contrast to recent efforts to map variance QTLs using un-replicated data which conflates the two [52]–[55] . To test if defense or growth traits may differentially affect the link between CV and mean , we also measured the variation in growth and defense chemistry [51] . As found in a previous analysis of the Arabidopsis transcriptome , stochastic variation showed a higher heritability than that for variation in the average phenotype . As found for the transcriptome , there were differences in the genetics controlling the stochastic variation and average phenotypes . In support of ecological/bet-hedging theory , defense chemistry showed more QTLs of larger effect for stochastic variance than those found for growth or primary metabolism . Importantly , the genetic variation within the organelle had a widespread effect on the stochastic variation in primary metabolism with discrete impacts that differed from the organelle effect on the average metabolome . Thus , natural variation has widespread effects on the stochastic variation of growth and metabolism involving both the nuclear and organellar genomes . Future work will identify if the genetic basis of the average and stochastic variation are caused by similar or dissimilar mechanisms .
To test if genetic variation affects stochastic noise in the metabolome and growth of the higher plant Arabidopsis thaliana , we used a previous analysis of quantitative variation of the average metabolism and growth within the Kas x Tsu RIL population [51] , [56] . A total of 559 metabolomic , 19 chemical defense and 5 growth traits were measured in this population with replicated independent experiments providing replication on both the average and standard deviation of each phenotype . Using this data , we obtained the coefficient of variance ( CV ) for each phenotype in each experiment for each RIL . This was done by dividing the standard deviation of the phenotype within an experiment by its mean within that same experiment . CV is an appropriate comparative measure of genotypic stochastic noise as it is a dimensionless measure of variation allowing us to perform the ensuing analysis [16] , [21] . All per line CV measures were compared to the previously published analysis of the average phenotypes for the same traits [51] , [56] . As previously found using the Arabidopsis transcriptome , the heritability for the metabolite CV was higher than that for the average metabolite accumulation ( Fig . 1A and S1 Table ) [24] . In addition to the metabolome , both growth and defense chemistry also showed increased heritability for per line CV in comparison to the average ( S1 Fig . ) . Comparing the heritability of per line CV and average across all the metabolites showed that there was no correlation between these two values ( Fig . 1B ) . Similarly , there is no correlation between mean and CV for the metabolites across all the RILs ( S2 Fig . ) . Thus , per line CV is not being driven simply by variation in the level of the average phenotype within this dataset but is instead an independent output of the genetic variation in comparison to the average metabolite accumulation . Similar to the transcriptome , the range of metabolite CV across the RILs was less than that found for the average metabolite accumulation ( Fig . 1D ) . The Kas x Tsu population is a reciprocal population that allows us to measure the relative contribution of the nuclear and organellar genomes to any resulting phenotypes by using the maternally inherited organellar genomes as a single marker [57] . Because these RILs are in their F10 generation due to bulking in our lab and all seed mothers for the RILs for this experiment were grown together and harvested at the same time , we are largely focusing on maternal effects due to the genetic variation in the organelles . Thus , we used a linear model to estimate the contribution of the organellar genome variation to heritability of per line CV across the metabolome . This showed that the organellar genomes contributed 5 . 4%±0 . 2% heritability with a max of 31% heritability for metabolites ( Fig . 1A and S1 Table ) . This organellar genome heritability for per line metabolite CV was significantly higher than that found for average metabolite accumulation ( Fig . 1A ) [51] . Again , there was no correlation between the heritability of per line CV and average driven by the organellar genome across the metabolites ( Fig . 1C ) . This suggests that as with the nuclear genome , the effect of the organellar genomic variation on CV is separate from that of the effect on average metabolite accumulation ( Fig . 1C ) . In contrast to the metabolome , the cytoplasm had similar heritable effects on the CV of growth and defense chemistry as that found for the average ( S1 Fig . ) . Thus , the genetic variation in the organelles of Arabidopsis can heritably influence per line CV of plant metabolism , growth , and defense chemistry . Using per line CV and average metabolite accumulation across all the RILs , we can obtain the genetic coefficient of variation across the population ( Population CV ) , which describes the range of variation for that trait across the RILs . Correlating range of variation across the population for CV and average using all the metabolites showed that there was a continuous range of variation in the relationship between population variation in mean and CV . To test if there might be some biological insight within these distributions , we focused on the metabolites whose population variation that were in the top 5% or bottom 5% of the metabolites for either mean or CV . This allowed us to define three groupings ( Fig . 2 ) . One grouping was characterized by metabolites where the population CV is in the top 5% of all metabolites but the variation of average for these same metabolites is within the bottom 5% ( Top left of Fig . 2 ) . This included lipids , such as Steric and Palmitic acid , as well as energy sources into lipid metabolism , like glycerol and Glucose-1-P . Contrastingly , a set metabolites that consist predominantly of amino acids and sugars , were in the bottom 5% of all metabolites for population variation in both CV and average ( Bottom left of Fig . 2 ) . This would suggest that these metabolites are constrained or robust within this population . There was also a set of metabolites whose average accumulation was within the top 5% of all metabolites yet their CV was not an outlier ( Right of Fig . 2 ) . This included stress inducible metabolites like Putrescine , Isonicotinc acid , Salicylic acid , Shikimic acid and Methionine ( Fig . 2 ) . These metabolites should be the more sensitive to micro-environmental variation in stress than the other compounds . The fact that these stress sensitive metabolites only have intermediate variation in CV within this population further suggests that we are measuring genetic diversity in CV rather than any micro-environmental effect . We obtained the average and per line CV for each metabolite for each RIL from the linear model used to estimate heritability . We used these values to map QTLs for both phenotypes across all 559 metabolites for all 271 RILs with fully replicated data . This analysis identified on average 3 QTL for 434 metabolites using the average accumulation and 1 . 75 QTL for 414 different metabolites using the per line CV ( Fig . 3 and S3 Table ) [51] , [56] . There was no observable correlation in the number of CV or average QTLs across the metabolites nor in the effect of overlapping QTLs ( S3 Fig . ) . This decrease in QTL identification for per line CV is similar to previous analysis using transcriptomic variation in a different Arabidopsis population [24] . The mean effect of each identified metabolite average QTL was 22% in comparison to 17% for metabolite CV QTL , which also agrees with what was previously found using the transcriptome ( Figs . 3 and 4 ) [24] . The fact that per line CV has higher heritability with fewer detectable QTL of lower effect size than the average phenotype suggests that per line CV likely has a more polygenic genetic basis than that controlling the average metabolite accumulation [58] , [59] . A comparison of the QTL maps across all the metabolites showed that the patterns of loci were not identical ( Fig . 4 ) . This suggested that there might be different loci controlling the average and CV of metabolite accumulation in these RILs . Overlapping the QTL hotspots identified using the average and CV metabolic phenotypes across all metabolites showed that this was in fact the case ( Fig . 5 ) . There were QTL hotspots specific for either the average or per line CV of metabolite accumulation . There were five hotspots statistically unique to per line CV . For example , the QTL on Chromosome II ( M . CV . II . 15 ) was entirely linked to per line CV in metabolite accumulation with no detectable effect on average metabolite accumulation ( Fig . 5 ) . Similarly , there were seven hotspots statistically significantly enriched only in average metabolite accumulation ( Fig . 5 ) . The three loci on chromosome I for average metabolite accumulation had the most specific effects on average ( M . AV . I . 50 , −63 and −83; Fig . 5 ) . There were also four loci that were hotspots for both average and per line CV of metabolite accumulation ( M . III . 51 , M . III . 64 , M . IV . 3 and M . IV . 72 ) . Thus , the genetics of per line CV and per line average metabolite accumulation can identify sets of genetic loci that include loci specific for one or the other trait . This suggests that stochastic variance of plant metabolism is a heritable genetic trait distinct from that of per line average . Several recent modelling studies had used predictive models of the metabolic grid and suggested that it was possible for stochastic noise within the metabolome to be constrained to specific regions of the grid [32] , [33] . To test if our empirical data shows if the CV QTLs have localized effects on metabolite CV as predicted from the models , we plotted the significant additive effects of each locus within a diagram of the metabolic grid ( S4 Fig . ) . These plots showed that the effects of some QTL on metabolite CV were typically localized to a relatively small region . At the extreme were loci that affected only specific nodes within the detectable primary metabolic grid , such as M . CV . V . 97 and M . CV . II . 16 ( S4 Fig . ) . In contrast to the predictions , there were a number of loci that had wide ranging effects scattered throughout the metabolic grid , such as M . CV . III . 51 , M . III . 64 and M . CV . IV . 72 ( S4 Fig . ) . These effects were both positive and negative within the same metabolome . For example , M . CV . III . 51 showed increased variance in succinate and xylose while decreased variance in spermidine , glycerate , glu-1-P and other metabolites ( S4 Fig . ) . Thus , in contrast to the modelling studies , it is possible for genetic loci to have wide ranging and opposing effects upon metabolome stochastic variance . To compare how per line CV loci differ across phenotypic classes , we next used per line CV and average for each RIL for growth and defense chemistry to map QTLs for these phenotypes . As for metabolites , this showed that the average phenotype found more QTLs for all traits than that found for per line CV ( 4 , 7 and 7 versus 2 , 1 and 1 for aliphatic glucosinolates , indolic glucosinolates and growth respectively ) ( Fig . 3 and Tables S4 and S5 ) . In contrast to the rest of the metabolome and transcriptome , the effect size of defense chemistry per line CV QTLs was larger than that for the QTLs affecting the average . The CV QTLs have a mean effect of 57 and 42% for aliphatic and indolic glucosinolates , in contrast to the average QTLs having a 40 and 20% effect respectively ( Fig . 6 and S3 Table ) [51] , [56] . Similarly , effect of the per line CV QTLs for growth was also higher than that for average growth , 21% effect versus 10% ( S3 Table ) [51] , [56] . In all growth and defense phenotypes , the distribution of effect sizes for the phenotypic per line CV was statistically higher than for the phenotypic average ( t-test , P<0 . 01 ) . It should be noted that all growth , defense , and metabolite phenotypes were measured on the same plants indicating that these differences are not likely due to different environments or conditions [51] , [56] . This increased effect size of QTLs for per line CV of growth and defense chemistry in comparison to that found for the metabolome suggests that the underlying genetics controlling the per line CV of growth and defense chemistry is structured differently between the traits . A comparison of QTL maps for the defense traits showed that the previously identified and validated GSL . AOP and GSL . Elong loci control both mean and per line CV for the aliphatic glucosinolate ( S5 Fig . ) [24] . The stochastic variation and mean accumulation of the aliphatic glucosinolates is controlled by the presence or absence of specific enzyme encoding genes in these loci that lead to pleiotropic effects on the glucosinolate regulatory network [24] , [60]–[63] . The GSL . AOP and Elong loci were also linked to suggestive hotspots ( P<0 . 1 ) in the average metabolome with no signature in the metabolome per line CV ( Fig . 5 ) . For aliphatic glucosinolates , there is also a per line CV hotspot near the previously validated MYB28 locus , a transcription factor , that also controls the glucosinolate regulatory network to affect stochastic variation of the pathway ( S5 Fig . ) [24] , [64]–[67] . In contrast to the CV analysis of the metabolome , there were no significant hotspots that were unique to defense chemistry per line CV ( S5 Fig . ) . Mapping per line CV of growth in comparison to average growth identified a number of average QTLs and only two CV loci for growth ( Figs . 5 and S5 ) . The growth QTL , GR . I . 19 was associated with variation in both average and per line CV of growth while the QTL , GR . III . 2 , was specific to per line CV in growth ( Tables S4 and S5 ) . There were no hotspot in the metabolome or defense chemistry data for the GR . I . 19 or the GR . III . 2 loci suggesting that the effect of these loci on the altered per line CV in growth was not having a detectable impact on metabolism Interestingly , only one average or per line CV growth locus ( GR . IV . 2 vs M . IV . 3 ) overlapped with any metabolomics locus in the entire analysis suggesting we identified different genetic loci for the two traits . Together , this shows that we can map loci for per line CV of growth , metabolism , and defense chemistry and identify loci specific to each trait . Thus , per line CV loci are genetically distinct for all three traits and not reflective of a global stochastic noise locus .
Studies on stochastic variation have difficulty discerning if the observed genetic effects on CV are truly via intrinsic processes . An alternative is that the loci could be reflecting genetic variation that alters the phenotypic plasticity in the presence of micro-environmental perturbations . We would argue that our data is more reflective of intrinsic stochastic variation for the following reasons . First , our experiment was conducted with complete randomization at all levels that should prevent any signature of local environmental structure in technical or biological replicates . Essentially all samples should be equally randomized across any micro-environmental variation . In support of this , diurnally responsive metabolites show all ranges of CV indicating that any effect of micro-diurnal variation on the sampling and CV estimation is minimal ( Fig . 2 ) [71] . Further supporting this is the observation that stress responsive metabolites are not showing elevated CV as would be expected if we were measuring plasticity in response to micro-environmental variation in stress . Secondly , the primary metabolites , secondary metabolites and growth were all measured on the same plants and as such should be exposed to the same micro-environmental variation . Yet the loci identified and genetic architecture of these traits is fundamentally different suggesting that we have mapped loci specific to each metabolic trait and not universal plasticity loci . Thirdly , there were no loci identified with structured global effects in metabolic CV as would be expected if there was the presence of systemic structured biological or technical error ( Figs . 7 and S5 ) . Supporting the absence of systemic sources of error came from randomizing the metabolomic data while maintaining the inherent structure . This analysis found that the maximal number of QTLs found was 53 which is only 9% of the 595 CV QTLs identified with the real data arguing against systematic error . Finally , we have previously used this same experimental set up to identify and validate that ELF3 specifically affects intrinsic stochastic noise [24] . Thus , we would argue that while some of our loci may be loci affecting plasticity to extrinsic variance , we have likely identified a number of loci that affect intrinsic stochastic variance within the metabolome and growth in a multi-cellular eukaryote . It will require vastly larger validation experiments to separate which loci are associated with intrinsic vs extrinsic stochastic variance . The link between genetic variation and differential stochastic noise in a phenotype has been predominantly studied in single celled organisms [16]–[20] , [72] . Additionally , in plants there are whole plant processes that rely on stochastic cell autonomous processes , such as flowering time [73] , [74] . This has generated some confusion over the potential for stochastic variation at the whole plant versus cell autonomous level . However , previous work showed that it was possible to identify whole plant stochastic events controlled by genetic polymorphisms buffered by HSP90 [13] , [14] , [75] . Within our analysis we mapped genetic variation that influenced the stochastic variation of plant growth as measured by the size of the whole rosette . Plant growth is a classical integrative higher-order phenotype like crop yield or disease susceptibility having complex underlying genetics [76] , [77] . Thus , it is possible to identify genetic loci that determine the level of stochastic variation in whole plant phenotypes . It remains to be seen if the underlying molecular mechanisms work in cell non-autonomous manners to control whole plant phenotypes or function as stochastic switches in cell autonomous manners that sum up to a whole plant result . Recent research is beginning to unveil the role of genetic variation within organellar genomes in influencing variation for a range of phenotypes from average metabolite accumulation to growth [51] , [56] . Further , only diversity in nuclear encoded genes like ELF3 have been linked to influencing stochastic variation within plants [24] . Thus , there has not yet been an identified link of the organellar genome variation to controlling different stochastic variation within any organism . Within our study , we found that genomic variation within the organelles lead to a significant impact on the stochastic variation of metabolites as measured by per line CV ( Fig . 7 ) . There was also a lesser impact on the defense metabolites and growth ( S6 and S7 Figs . ) . The variation within the organellar genome influenced stochastic variation of primary metabolism differently than average metabolite accumulation . Thus , the organelle genome influences stochastic variation at all phenotypic levels and the CV effects can be separated from the effects on the average phenotypes and these effects are due to genes within the organellar genomes . It has been hypothesized that defense related phenotypes benefit from having elevated levels of stochastic variation that generate a bet-hedging-like mechanism whereby a single genotype samples a wider phenotypic range . This can then lead to increases in evolutionary stability of the defense mechanism . Within this experiment , defense metabolites had numerous lines of evidence indicating that they had a higher per line CV and more genetic variation in per line CV than is found for primary metabolites in agreement with this theory . First , defense metabolites have a wider population level variance of per line CV than that found for the other metabolites ( aliphatic glucosinolates 1 . 5±0 . 3 , indolic glucosinolates 0 . 8±0 . 3 and primary metabolites 0 . 5±0 . 1 [average ± S . E . of population CV for per line CV] ) ( S2 Table ) . Additionally , we identified more per line CV QTLs for each defense metabolite than for the other metabolites ( Fig . 5 ) . Finally , for each identified QTL controlling per line CV , the mean effect for defense metabolites was twice as large as that found for the other metabolites ( 57% effect for aliphatic glucosinolates , 42% effect for indolic glucosinolates and 22% effect for primary metabolites ) ( Fig . 6 ) . Taken together , there is a higher level of genetically programmed stochastic variance in glucosinolate defense metabolites in comparison to primary metabolites . Thus , the genetic networks and natural variation influencing defense metabolism in Arabidopsis may be structured to enable higher levels of stochastic variation possibly to mediate bet-hedging interactions within the environment [28] , [29] . Within this study , we show that it is possible to identify genetic loci in both the nuclear and organelles that lead to altered stochastic variation in all measured phenotypes from individual metabolites to whole plant growth . Further , these loci differ from trait to trait , suggesting that we are not identifying generic variance loci as might be expected if they were affecting global mechanisms like HSP90 . Instead , these CV loci affect specific genetic networks that are distinct for each trait . This suggests that there may be stochastic specific loci for each plant trait . For instance , numerous natural and induced mutant screens and surveys have been conducted in Arabidopsis to determine the genes controlling the phenotypic average [78]–[80] . Similar large scale approaches have been conducted in numerous other organisms focused on phenotypic averages [31] , [81] , [82] . While these have provided great advances in our understanding of biology , it raises the question of what would happen if we repeat these screens and surveys to identify genetic variation controlling stochastic noise in phenotypes . Would we identify the same genes or would we begin to identify a large suite of previously unknown genes that control stochastic variation rather than phenotypic average ? This indicates there is a need for additional experiments focused on stochastic variation within multi-cellular organisms to explore a new avenue of organismal biology .
To directly estimate the CV for each individual metabolites accumulation as a separate phenotype within the Kas x Tsu RIL population [51] , [56] , we utilized two independent metabolomics experiments in which 316 lines had been measured in duplicate within each experiment [51] , [56] . Within each experiment , the 316 lines were planted in randomized complete blocks and all blocks within all experiments were independently randomized . This greatly diminishes any potential for correlated errors in the analysis . Additionally , the metabolomics samples were also randomized prior to injection within the block structure . Again all randomization was independent across blocks for the metabolomics . Only 559 metabolites were measured in all four samples of the previous experiment and we focused solely on these signals to maximize our power to measure metabolite CV [51] , [56] . To measure growth and defense compound CV , we obtained the raw data where the plants had also been measured for daily growth ( 5 growth phenotypes ) and glucosinolate accumulation ( 19 glucosinolate phenotypes ) [51] , [56] . For each phenotype , metabolite and growth , we utilized the absolute phenotypic values to measure the CV for each phenotype separately for each experiment using σ/µ [16] , [21] , [83] , thus providing two independent biological replicate measures of CV for each phenotype . The use of CV as a direct phenotype has previously been used in a number of instances . By measuring the within line CV as a phenotype for the Kas x Tsu population allows us to then utilize CV as a direct measurement of stochastic variation as a phenotype . The level of per line replication for the array data does not support the use of Levene's variance tests or measures . Additionally , all lines were planted and harvested within a randomized complete block design at all stages thus limiting any potential technical bias to generate these observations [41] , [84] . Similarly , the metabolomics analysis was conducted with mixed internal standards run approximately every 20 samples to normalize all of the runs to minimize any potential technical error from the instrument [85]–[87] . For estimating broad-sense heritability , we utilized the independent measures of CV directly as a phenotypic measure . All RIL lines were represented in every block in both experiments creating a perfectly balanced randomized complete block design . All phenotypic data was used to calculate estimates of broad-sense heritability ( H ) for each phenotype as H = σ2g/σ2p , where σ2g was estimated for both the RIL genotypes and cytoplasmic genotypes and σ2p was the total phenotypic variance for a trait [88] . The ANOVA model ( Line heritability Model ) for each metabolite phenotype in each line ( ygmeb ) was: where c = the Kas or Tsu cytoplasm; g = the 1…316 for the 316 RILs , e = experiment 1 or 2 . This allowed cytoplasmic effects to be directly tested in the C term and each RIL genotype ( G ) nested within the appropriate cytoplasmic class , either Kas or Tsu . Experiment was treated as a random term within the model to better parse the variation . All resulting variance estimates , P-values and heritability terms are presented ( S1 Table ) . σ2g for RIL was pulled from the Gg ( C c ) term while σ2g for cytoplasmic variation was pulled directly from the C cMm term . We used mean CV values for each RIL for further analysis as we had a randomized complete block design with no missing lines ( S2 Table ) . We used the previously reported genetic map for these lines of the Kas × Tsu RIL population [56] , [57] . To detect CV QTLs , we used the average CV per phenotype per RIL across all experiments ( S2 Table ) [56] , [57] . For QTL detection , composite interval mapping ( CIM ) was implemented using cim function in R/qtl package with a 10 cM window . Forward regression was used to identify three cofactors per trait . To control for genome-wide false positive rates , declaration of statistically significant QTLs was based on permutation-derived empirical thresholds using 1 , 000 permutations for each mapped trait which yielded a range of LOD significances of 1 . 8–3 . 5 to call significant QTLs . In addition to setting a significance threshold , this approach also randomizes the genotype-to-phenotype link to establish a false positive rate . To be conservative , QTLs with a LOD score above 2 were considered significant for further analysis [89] , [90] . Composite interval mapping to assign significance based on the underlying trait distribution is robust at handling normal or near normal trait distributions [91] , as found for most of our phenotypes . The define peak function implemented in R/eqtl package was used to identify the peak location and one-LOD interval of each significant QTL for each trait [92] . The effectscan function in R/qtl package was used to estimate the QTL additive effect [93] . Allelic effects for each significant QTL are presented as percent effect , by estimating for each significant main effect marker ( S3 Table ) . QTL clusters were identified using a QTL summation approach where the position of each QTL for each trait was plotted on the chromosome by placing a 1 at the peak of the QTL . This was then used to sum the number of traits that had a detected QTL at a given position using a 5cM sliding window across the genome [94] . The QTL clusters identified defined genetic positions that were named respective to their phenotypic class and genetic positions with a prefix indicating the phenotype followed by the chromosome number and the cM position . For example , M . CV . II . 16 indicates a CV metabolomics QTL hotspot on chromosome II at 16 cM . The cluster analysis was conducted separately for metabolomic , defense chemistry and growth phenotypes . To further assess the potential of structured technical or biological variation to influence our analysis , we conducted a permutation analysis wherein we randomized the line to metabolome links within each of the four randomized blocks . This maintains any correlative structure between the metabolites within a metabolomic sample that may have been caused by structured technical or biological error . We then recalculated CV and mean within each RIL using the randomized phenotype data and used this to re-conduct the entire QTL analysis as described above . 100 permutations of the entire dataset identified a maximum of only 53 metabolomic CV QTL identified across the 559 metabolites in any given permutation which lead to no hotspots being identified . This suggests that the observed hotspots are not caused by structured error within the metabolomics samples . To directly test the additive effect of each identified QTL cluster , we used an ANOVA model containing the markers most closely associated with each of the significant QTL clusters as individual main effect terms . For each metabolite the average accumulation in lines of genotype g at marker m was shown as ygm . The model ( Additive Model ) for each metabolite in each line ( ygm ) was: where g = Kas ( 1 ) or Tsu ( 2 ) ; m = 1 , … , 11 . The main effect of the markers was denoted as M involving 15 markers ( m ) . The cytoplasmic genome was included as an additional marker to test for cytoplasmic genome effects . We independently tested the average metabolite accumulation and CV of each metabolite as a separate phenotype with the appropriate model using lm function implemented in the R/car package , which returned all P values , Type III sums-of-squares for the complete model and each main effect . The results using the average metabolite accumulation are presented ( S4 Table ) separately from those for the CV of metabolite accumulation ( S5 Table ) . QTL main-effect estimates ( in terms of allelic substitution values ) were estimated for each marker [93] , [95] . The same analysis was conducted for the aliphatic glucosinolates , indolic glucosinolates and growth except that these phenotypes only had 9 loci instead of 10 ( Tables S4 and S5 ) . There is no significant single marker or pairwise segregation distortion in this population indicating that the model is balanced for all markers [57] . To test directly for epistatic interactions between the detected QTLs , we conducted an ANOVA using the pairwise epistasis model . We used this pairwise epistasis model per metabolite because we had previous evidence that RIL populations have a significant false negative QTL detection issue and wanted to be inclusive of all possible significant loci [49] . Within the model , we tested all possible pairwise interactions between the markers . For each phenotype , the average value in the RILs of genotype g at marker m was shown as ygm . The model ( Pairwise epistasis model ) for each metabolite in each line ( ygm ) was: where g = Kas ( 1 ) or Tsu ( 2 ) ; m = 1 , … , 14 and n was the identity of the second marker for an interaction . The main effect of the markers was denoted as M having a model involving 15 markers . The cytoplasmic genome was included as an additional single-locus marker to test for interactions between the cytoplasmic and nuclear genomes . We independently tested the average metabolite accumulation and CV of each metabolite as a separate phenotype with the appropriate model using lm function implemented in the R/car package , which returned all P values , Type III sums-of-squares for the complete model and each main effect . The results using the average metabolite accumulation are presented ( Tables S6 and S7 ) separately from those for the CV of metabolite accumulation ( Tables S8 and S9 ) . Significance values were corrected for multiple testing within a model using FDR ( <0 . 05 ) . The main effect and epistatic interactions of the loci were visualized using cytoscape . v2 . 8 . 3 with interactions significant for less than 10% of the phenotypes were excluded from the network analysis [44] , [96] . The 10% threshold was chosen as an additional correction for multiple testing to provide a more conservative image of the network . The same analysis was conducted for the aliphatic glucosinolates , indolic glucosinolates and growth except that these phenotypes only had 9 loci instead of 10 ( Tables S6 to S9 ) . There are no pairwise locus segregation distortions within this population showing that the genotypes in this analysis are balanced [57] . | Systems biology is largely based on the principal that the link between genotype and phenotype is deterministic , and , if we know enough , can be predicted with high accuracy . In contrast , recent work studying transcription within single celled organisms has shown that the genotype to phenotype link is stochastic , i . e . a single genotype actually makes a range of phenotypes even in a single environment . Further , natural variation within genes can lead to each allele displaying a different phenotypic distribution . To test if multi-cellular organisms also display natural genetic variation in the stochastic link between genotype and phenotype , we measured the metabolome , growth and defense metabolism within an Arabidopsis RIL population and mapped quantitative trait loci . We show that genetic variation in the nuclear and organeller genomes influence the stochastic variation in all measured traits . Further , each trait class has distinct genetics underlying the stochastic variance , showing that there are different mechanisms controlling the stochastic genotype to phenotype link for each trait . Further work is necessary to identify the mechanisms underpinning the stochastic nature of the genotype to phenotype link . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"biotechnology",
"systems",
"biology",
"genome",
"complexity",
"plant",
"science",
"genomics",
"phenotypes",
"plant",
"genomics",
"genetics",
"biology",
"and",
"life",
"sciences",
"population",
"genetics",
"plant",
"genetics",
"computational",
"biology",
"evolutionary",
"biology",
"population",
"biology",
"plant",
"biotechnology"
] | 2015 | Genetic Variation in the Nuclear and Organellar Genomes Modulates Stochastic Variation in the Metabolome, Growth, and Defense |
In tropical regions , protozoan parasites can cause severe diseases with malaria , leishmaniasis , sleeping sickness , and Chagas disease standing in the forefront . Many of the drugs currently being used to treat these diseases have been developed more than 50 years ago and can cause severe adverse effects . Above all , resistance to existing drugs is widespread and has become a serious problem threatening the success of control measures . In order to identify new antiprotozoal agents , more than 600 commercial agrochemicals have been tested on the pathogens causing the above mentioned diseases . For all of the pathogens , compounds were identified with similar or even higher activities than the currently used drugs in applied in vitro assays . Furthermore , in vivo activity was observed for the fungicide/oomyceticide azoxystrobin , and the insecticide hydramethylnon in the Plasmodium berghei mouse model , and for the oomyceticide zoxamide in the Trypanosoma brucei rhodesiense STIB900 mouse model , respectively .
The Protozoan parasites of the genera Plasmodium spp . , Leishmania spp . , Trypanosoma brucei spp . and Trypanosoma cruzi , are the disease causative agents threatening entire populations in mainly resource poor countries around the world . Malaria , due to infection with Plasmodium spp . , is one of the most devastating diseases in developing countries , with 216 million cases in 2010 , causing an estimated 655 , 000 deaths per year [1] . Other recent estimates assume up to 1 . 2 million deaths per year [2] . For the treatment of malaria several highly active drugs are available , like chloroquine , quinine , mefloquine , atovaquone , artesunate , and their analogs . Thus , malaria is often not included in the list of the neglected tropical diseases . Unfortunately , significant resistance to almost all of these drugs has developed; even to the “last resort” artemisinin-derivatives , first cases of delayed clinical efficacy have been reported [3] . Recently , large libraries from pharma companies have been screened against protozoan parasites and some interesting hits [4] , [5] , [6] , [7] have been found , especially against malaria with the spiroindolones currently undergoing clinical evaluation [8] , [9] . Most of the promising compounds in the development pipeline are in a rather early clinical stage , so that a high failure rate is expected [10] . Considering the rapid development of resistance , and the challenges seen with the development of malaria vaccines [11] , a continuous refilling of research pipelines with compounds in preclinical/clinical evaluation will be necessary , for the long term perspective . Therefore new compounds for resistance management would be highly desirable , even if they might not show the same remarkably high activity levels as the recently promoted peroxide candidates like OZ439 [12] . In addition , the global malaria agenda has shifted from the mere control of clinical cases to malaria elimination and eventually eradication urgently requiring transmission blocking agents [13] . Human African trypanosomiasis ( HAT ) , also known as sleeping sickness , is caused by infections of T . b . rhodesiense and T . b . gambiense . Populations living in remote rural areas of sub-Saharan Africa are at risk of acquiring HAT . The disease burden in 2000 was estimated at 1 . 3 Mio DALYs ( Disability-Adjusted Life Years ) and the estimated number of cases up to 70 , 000 in 2006 [14] . In recent years the public health situation has improved due to increased monitoring and chemotherapy , resulting in the decrease of reported HAT cases to approximately 10 , 000 [15] . Only 4 drugs are currently registered as HAT treatment . Pentamidine and suramin are used to treat the hemolymphatic stage ( stage 1 ) of the disease , while melarsoprol and eflornithine ( DFMO ) are used in stage 2 of the disease when the parasites have invaded the central nervous system ( CNS ) and which is lethal if untreated . The available drugs are unsatisfactory due to cost , toxicity , poor oral bioavailability , long treatment and lack of efficacy . Melarsoprol is highly toxic , and up to 5% of the second stage patients treated with melarsoprol die of a reactive encephalopathy . Eflornithine treatment is expensive and logistically difficult; it requires four daily intravenous infusions over fourteen days . Recently the eflornithine-nifurtimox combination therapy ( NECT ) was introduced [16] . The requirement of intravenous administration although reduced to a quarter of injections as compared to monotherapy is still a limitation , with a need for new and more easily administrable drugs . Trypanosoma cruzi infection elicts Chagas disease and is an important public health problem causing approximately 14 , 000 deaths and 0 . 7 Mio DALY annually [17] . Treatment options are limited due to toxicity of available drugs , parasite resistance , and poor drug activity during the chronic phase of the disease . Currently there are two medications being used to treat Chagas disease , nifurtimox and benznidazole [18] . Severe toxicity and long treatment requirements are associated with both drugs [19] . Therefore new medications are badly needed for treating this disease especially in its chronic phase . Leishmaniasis causes approximately 50 , 000 deaths and 2 . 1 Mio DALY annually [20] . It threatens about 350 million people around the world and 12 million people are believed to be infected , with 1–2 million estimated new cases every year [21] . Widely used medications are still based on i . v . application of antimony compounds like stilbogluconate , resulting in severe side effects . More modern , but also more expensive medications are liposomal amphotericin B , miltefosine , and paromomycin [22] . Thus new affordable and effective therapies are urgently needed to combat these disastrous diseases . Registration requirements for agrochemicals are in some aspects even more stringent than for pharmaceuticals , as side effects that are tolerated for drugs against many life threatening diseases , are not acceptable for agrochemicals that potentially could enter the food chain [23] , [24] , [25] . As a consequence , all commercialized agrochemicals must go through broad toxicological profiles including e . g . chronic and reprotoxicological studies in different mammalian species , covering at least part of the preclinical studies required for drug development . Furthermore , agrochemicals are highly optimized on agrochemical pest targets with often good selectivities in mammals and excellent temperature and storage stability . Another interesting feature of commercial agrochemicals is the very low production cost of only a few cent/g , as the compounds are produced in highly optimized processes on the multi-ton scale . Surprisingly , these aspects have not led to a systematic evaluation of agrochemicals for pharmacological use so far [26] . Here we present data of over 600 commercial agrochemicals which have been systematically tested for the first time for their antiparasitic activity .
A library of over 600 compounds ( for a list of CAS-numbers and common names of the tested agrochemicals see Supporting Information S1 ) , that are or have been active ingredients in commercial agrochemical products , has been compiled from the BASF compound depository and was dissolved in DMSO stock solutions in a concentration of 10 mg/ml . These samples were then further diluted according to the requirements of the assays . The structural integrity of the dissolved samples has been confirmed subsequently by LCMS-analysis . All work was conducted in accordance to relevant national and international guidelines . The in vivo efficacy studies were approved by the veterinary authorities of the Canton Basel-Stadt . The in vivo studies were carried out under license No . 1731 and license No . 739 of the Kantonales Veterinäramt , CH-4025 Basel , Switzerland adhering to the Tierschutzverordnung from 23 . 04 . 2008 ( based on the Tierschutzgesetz from 26 . 12 . 2005 ) .
38 agrochemicals with sub-µM activity on T . cruzi were identified , many of which being azoles with P450-inhibiting activity ( Figure 2 ) . P450-monoxygenases have been discussed before as targets against T . cruzi , especially the sterol 14α-demethylase [49] . The standard drug benznidazole ( LD50 rat p . o . not available ) [50] , [51] has an IC50 of 1871 nM in this assay . Ipconazole ( LD50 rat p . o . 888 mg/kg ) , has an IC50 of 3 . 0 nM , the most active agrochemical against T . cruzi . It is a fungicide used predominantly in seed dressing . The tested material is , like the commercial material , racemic and a mixture of diastereomers , therefore an enantiopure isomer could potentially have even higher activity . Difenoconazole ( LD50 rat p . o . 1453 mg/kg ) , a broad spectrum and systemic fungicide , showed an IC50 value of 7 . 4 nM . This commercial agrochemical is again a racemic diasteromeric mixture and could therefore also have intrinsically higher activity as a pure isomer . Clotrimazole ( 14 nM ) , and viniconazole [52] ( 26 nM ) , are two azole drugs used against fungal skin infections , that have also been discussed as agro fungicides and therefore have been tested in this screen . As they have a complete pharmacological dossier they might also be interesting drug candidates . Zoxamide ( LD50 rat p . o . >5000 mg/kg ) , a broadspectrum oomyceticide used in fruits and vegetables , showed 27 nM activity . It is sold and was tested as a racemate . Its mode of action against oomycetes is the inhibition of microtubule formation . Pyridaben ( 30 nM ) , and tolfenpyrad ( 55 nM ) , are insecticides/acaricides inhibiting the complex 1 in the mitochondrial electron transport chain . A number of further azole fungicides showed activities below 100 nM including metconazole 31 nM , tebuconazole 36 nM , bitertanol 35 nM , climbazole 55 nM , prochloraz 69 nM , hexaconazole 73 nM , and fenapanil 99 nM . Further agrochemicals with high activity in this assay were penconazole ( 130 nM ) , epoxyconazole ( 136 nM ) , imazalil ( 148 nM ) , propiconazole ( 160 nM ) , fenarimol ( 193 nM ) , fluquinconazole ( 199 nM ) , picoxystrobin ( 248 nM ) , cyproconazole ( 257 nM ) , myclobutanil ( 374 nM ) , tetraconazole ( 478 nM ) , and pyrifenox ( 491 nM ) . In spite of the excellent in vitro activity initial experiments in a T . cruzi mouse model did so far not show in vivo efficacy for selected hits ( personal communication Nazaré Soiro ) . Against L . donovani only two agrochemicals showed sub-µM activity ( Figure 3 ) . The standard miltefosine ( LD50 rat p . o . 246 mg/kg ) showed in this assay an IC50 value of 250 nM . Zoxamide ( LD50 rat p . o . >5000 mg/kg ) showed an IC50 of 250 nM . The oomyceticidal compound has been discussed in the T . cruzi section . Tolylfluanid ( LD50 rat p . o . >5000 mg/kg ) resulted in an IC50 value of 861 nM . It is a protective fungicide and oomyceticide with presumed thiol conjugating activity . Other agrochemicals with moderate activity against L . donovani were flocumafen ( 2451 nM ) , dimoxystrobin ( 3248 nM ) , bromofenoxin ( 3839 nM ) , cyhexatin ( 4517 nM ) , and cyazofamid ( 4988 nM ) . Due to the split of most life science companies into their agro- and pharma branches in the 1990s , the companies active in agrochemistry have not been involved in the recent screening activities to identify new drugs against infectious tropical diseases , even though agrochemicals might have a high potential to yield interesting hits for these applications . In this cooperation between industrial and public partners , it was shown for several commercial agrochemicals that they are highly active against some of the most important pathogens of infectious tropical diseases . Interestingly as anticipated , several of the oomyceticides ( strobilurins against P . falciparum , zoxamide against T . b . rhodesiense and L . donovani ) were active against these protozoans , but also other agrochemicals ( e . g . hydramethylnon against P . falciparum; azoles like iproconazole against T . cruzi ) showed very interesting activities . Exemplified by one of the major commercial agrochemicals , the fungicide azoxystrobin , as well as for the insecticide hydramethylnone , the reduction of parasitemia , and significant life extension for P . berghei infected mice was achieved . For zoxamide , an effect against T . brucei in the mouse model was also demonstrated . This successful in vitro– in vivo transfer without galenic optimization could not be taken for granted , as these agrochemicals have not been optimized for mammalian pharmacokinetics . There is still a high probability that the identified hits in the end might not be suitable for human use , as there are still several hurdles to overcome . However , the results of this highly focussed and relatively low input approach are more promising than could have been hoped for . It is especially noteworthy , that the screen of less than 700 agrochemical resulted in e . g . 24 new sub-µM hits against P . falciparum , compared to 4 new sub-µM hit in over 2687 recently tested commercial drugs ( excluding known antimicrobial and anti-cancer a . i . ) [54] , [55] . This clearly demonstrates that agrochemistry can be a very interesting and so far untapped source of new leads , and maybe even drug candidates , against protozoal diseases . It would also be very interesting to screen commercial agrochemicals against the pathogens of other neglected diseases , like schistosomes , nematodes , food borne trematodes , diarrhoeal amoebas and also tropical bacterial pathogens , for which good antibiotic cures are missing . These studies are still to be done . | Even though agrochemistry and infectious disease control have the same principle goal – the suppression of harmful organisms without harming human health and the environment – there have been only very limited activities to exploit this overlap for the development of new antiinfectious drugs so far . In this study and for the first time , over 600 commercial agrochemicals were systematically screened against the infectious pathogens causing malaria , sleeping sickness , Chagas disease and leishmaniasis . Many highly active compounds with known low mammalian toxicity were identified in cell based assays , and the activity of some of them could even be confirmed in first animal model studies . Further expansion of this concept to other pathogens and the examination of analogues of the identified hits , potentially available from agrochemical companies , would allow for a very efficient source of novel drug candidates . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"and",
"Discussion"
] | [
"plasmodium",
"falciparum",
"medicine",
"biotechnology",
"agrochemicals",
"infectious",
"diseases",
"chagas",
"disease",
"tropical",
"diseases",
"(non-neglected)",
"african",
"trypanosomiasis",
"leishmaniasis",
"neglected",
"tropical",
"diseases",
"malaria",
"agriculture"
] | 2012 | Agrochemicals against Malaria, Sleeping Sickness, Leishmaniasis and Chagas Disease |
Increasing amounts of sequence data are becoming available for a wide range of non-model organisms . Investigating and modelling the metabolic behaviour of those organisms is highly relevant to understand their biology and ecology . As sequences are often incomplete and poorly annotated , draft networks of their metabolism largely suffer from incompleteness . Appropriate gap-filling methods to identify and add missing reactions are therefore required to address this issue . However , current tools rely on phenotypic or taxonomic information , or are very sensitive to the stoichiometric balance of metabolic reactions , especially concerning the co-factors . This type of information is often not available or at least prone to errors for newly-explored organisms . Here we introduce Meneco , a tool dedicated to the topological gap-filling of genome-scale draft metabolic networks . Meneco reformulates gap-filling as a qualitative combinatorial optimization problem , omitting constraints raised by the stoichiometry of a metabolic network considered in other methods , and solves this problem using Answer Set Programming . Run on several artificial test sets gathering 10 , 800 degraded Escherichia coli networks Meneco was able to efficiently identify essential reactions missing in networks at high degradation rates , outperforming the stoichiometry-based tools in scalability . To demonstrate the utility of Meneco we applied it to two case studies . Its application to recent metabolic networks reconstructed for the brown algal model Ectocarpus siliculosus and an associated bacterium Candidatus Phaeomarinobacter ectocarpi revealed several candidate metabolic pathways for algal-bacterial interactions . Then Meneco was used to reconstruct , from transcriptomic and metabolomic data , the first metabolic network for the microalga Euglena mutabilis . These two case studies show that Meneco is a versatile tool to complete draft genome-scale metabolic networks produced from heterogeneous data , and to suggest relevant reactions that explain the metabolic capacity of a biological system .
Metabolic knowledge is crucial to understand physiology and biotic interactions . Supported by an unprecedented rise of sequencing technologies , the last decade saw the increasing understanding of metabolic capacities using genomic knowledge . In particular , in 2010 , Thiele and Palsson [1] described a general protocol enabling the reconstruction of high-quality metabolic networks , and several approaches have since been proposed to automate this process [2–4] . These methods rely primarily on two distinct steps . First , they provide automatic reconstructions of networks , called draft metabolic networks [5 , 6] , and in a second step fill the gaps of the draft networks . To this end , reference databases of metabolic reactions are used to check whether adding reactions to networks allows compounds of interest to be produced from given growth media . Identifying these missing reactions constitutes the so-called gap-filling problem ( n . b . considering the diversity of compound producibility and network consistency definitions , we here refer to a family of gap-filling problems rather than a single one ) [2] . Several approaches exist to solve gap-filling problems and to select , within databases , those reactions that must be added to the draft network to restore its consistency and a metabolic behavior . Reactions may be chosen to optimize a graph-based criterion [7] , or to optimize a linear score modelling the quantitative metabolic production of the system , as in the GapFill tool [8] , and its derivative fastGapFill [9] . Some approaches also integrate complementary knowledge such as taxonomic information [10] or compartment modularity [11] . More generally , the selection of reactions may be performed by optimizing a linear score modelling the consistency of a network with phenotypic knowledge , i . e . experimental flux data [12] or growth/no growth results [13] . Lastly , some tools combine several of the previously mentioned approaches . An example of these is MIRAGE , which selects reactions in a database in order to maintain biomass producibility with respect to a score based on co-expression and taxonomic distance between the target species and the species for which enzymes were evidenced [14] . The approach presented in [15] is similar although based on a different definition of producibility . In Table in S1 Table , we report the main characteristics of such methods in terms of required input data , the technological platform to be run , and examples of applications . Together these examples illustrate that methods to reconstruct metabolic networks have been very fruitful . However , as noted in [16] , most published genome-scale metabolic networks ( GEMs ) concern either prokaryotic or eukaryotic organisms for which genomic and physiological knowledge results from years of intensive studies . Indeed , GEM reconstruction is very sensitive to both genome annotation and the availability of complementary knowledge . It cannot , for instance , take into account genes of unknown function , which are common in incomplete or roughly annotated genomes . Nowadays , next generation sequencing ( NGS ) technologies are commonly employed to study strains and species distantly related to common model organisms . Draft metabolic networks based on these technologies are frequently quite degraded compared to those for standard model organisms . For instance , when comparing the number of reactions in the BioCyc repository [17] version 19 . 5 , we noticed that the 7 , 296 automatically reconstructed bacterial networks ( Tier 3 ) contained on average 8% fewer reactions than the 27 curated bacterial metabolic networks contained in the manually curated repositories ( Tier 1 & Tier 2 ) . For the sake of illustration , let us introduce two examples of organisms with a complex evolutionary history and recently studied using NGS technologies . Euglena mutabilis is a photosynthetic protist and important primary producer in acidic aquatic environments . Despite the crucial role of E . mutabilis in these ecosystems and the fact that it has often been considered as an indicator species for acid mine drainages ( AMDs ) , this organism has only been poorly described so far , in contrast to another species of the same genus , Euglena gracilis . The available data for E . mutabilis consists in assembled transcript sequences obtained from de novo transcriptomics and metabolomics experiments previously published in [18] and in [19] , respectively . This sparse dataset prevented us from using most of the tools described above to construct a metabolic network: the absence of a sequenced genome for the Euglena genus and the fact that this genus is not closely related to any common model organism rendered taxonomy-based methods of network reconstruction unusable [10 , 11] . E . mutabilis is difficult to cultivate in controlled conditions and to obtain as clonal cultures , preventing the use of phenotype-based tools [12 , 13] . The family of tools [8 , 9 , 14] that could be used here are of functional nature . Another application for gap-filling methods has emerged as a tool for studying the coexistence of organisms living in communities . As an example , Candidatus Phaeomarinobacter ectocarpi is a symbiotic bacterium associated with Ectocarpus siliculosus . Its genome and a draft of its metabolism could be produced [20] . Its host E . siliculosus has been studied for a longer time , and a functional metabolic network was reconstructed to explain the production of characteristic compounds of its metabolic profile [21] . Additional transcriptomic datasets were used in this study to identify 1 , 125 internal or external compounds produced by at least one reaction of E . siliculosus for which the corresponding enzyme was transcribed . Only 317 compounds could be produced according to the E . siliculosus draft network . In the framework of systems ecology [22] , a natural question is whether the symbiotic bacterial network can resolve some of this non-producibility . This issue can be rephrased as follows: how can the Candidatus Phaeomarinobacter ectocarpi metabolic network be used to fill gaps in the E . siliculosus draft GEM ? As above , this issue is of functional nature , and can be addressed only with functional gap-filling methods . Let us point out , however , that applying functional GEM gap-filling techniques [8 , 9 , 14] to organisms distantly related to common model organisms raises several problems . A first problem is related to the determination of the biomass reaction of the system . This reaction is often copied from well-established model organisms and therefore cannot capture all of the characteristics of the studied organism , especially when dealing with extremophiles . Shortcomings in the determination of an adequate biomass reaction lead to a second problem , that is , the determination of the boundary compounds , dead-end metabolites , and cofactors in the system . These may be hard to characterize from experiments or literature , despite their strong potential impact on the capacity of the system to produce biomass according to stoichiometry-based formalisms [8] . In particular , the score-based methods mentioned above depend on the stoichiometric balance of metabolic reactions , a criterion which may be prone to errors , especially with respect to cofactors and when using large-scale databases of metabolic reactions [2] . As a natural consequence , we advocate the need of GEM gap-filling techniques suitable for newly developed model organisms , in particular those with a complex evolutionary history and/or living in extreme environments for which phenotypic data are lacking . This study reformulates the gap-filling problem as a qualitative combinatorial ( optimization ) one . We introduce the tool Meneco ( Metabolic Network Completion ) that solves this problem , using Answer Set Programming ( ASP ) , a declarative programming paradigm including SAT-based solving technologies . Meneco considers reactions as achievable only if all their reactants are available , either as nutrients or provided by other metabolic reactions . Starting from given nutrients ( e . g . growth medium ) , referred to as seeds , this tool computes their scope defined as all the metabolites that can be synthesized from them using a graph-based approach . For metabolic network gap-filling , a database of metabolic reactions is queried to look for minimal sets of reactions that can restore the observed bio-synthetic behaviour ( i . e producibility of target metabolites ) . The Meneco tool was included in a pipeline implemented to construct EctoGEM , a metabolic network for the brown algal model E . siliculosus . The analysis of EctoGEM highlighted several interesting biochemical reactions , shedding light on the organization and evolution of some primary metabolic pathways of photosynthetic organisms [21] . In the present work , the case for Meneco as an important tool for hypothesis generation is further supported by new observations related to a benchmark of networks on a model organism and two case studies . First , this study simulates different degrees of manual curation using the model Escherichia coli [1] . For this purpose , 3 , 600 metabolic networks were generated from randomly degraded E . coli metabolic networks . In this benchmark , when the reference database used for completion was the real-case study Metacyc , Meneco outperformed the GapFill , the fastGapFill and the MIRAGE algorithms in terms of performance or accuracy . On a larger benchmark of 10 , 800 metabolic networks , our analysis suggests that Meneco is functionally relevant by identifying all essential reactions for more than 95% of the degraded networks . We advocate that the identification of such essential reactions is a key step towards the understanding of the metabolic capabilities of the species of interest , because they are related to key enzymes which , when removed , most likely prevent the viability of the species . Our results show that , when focusing on networks with a 10% degradation rate , the Meneco tool is able to restore the functionality of the network in 82% of cases . This suggests that Meneco is an important tool to study metabolic networks produced for organisms distantly related to common model organisms . In our first case study we use Meneco to assess the capability of the EctoGEM metabolic network to exchange metabolites with Candidatus Phaeomarinobacter ectocarpi , the aforementioned symbiotic bacterium associated with E . siliculosus . Combining the metabolic capacities of the draft GEM of E . siliculosus with those of the bacterial network enabled the in-silico production of 83 previously non producible algal targets . All of them were studied in detail allowing us to put forward hypotheses on possible exchanges between both organisms . Our second case study presents the first metabolic network for E . mutabilis , based on transcript sequences assembled from previously published transcriptomic and metabolomic data [18] [19] . In order to complete this draft with Meneco , we selected a set of targets from the list of metabolites that E . mutabilis can accumulate or secrete in minimum mineral medium [19] . Except for cobalamine , a cofactor that is not produced by this organism but is required for methionine synthesis , E . mutabilis can grow on a strictly mineral medium and is able to synthesize all the basic components of its biomass from mineral compounds only . Gaps in the draft network were filled iteratively with Meneco , using for each iteration a different subset of the 72 targets to solve the problem of cycles and circular dependencies . We thus obtained a network which was functional in Flux Balance analysis ( FBA ) for the photosynthetic production of biomass and excreted metabolites .
Our results suggest that the topological parsimony criterion used in Meneco provides a good trade-off in terms of scalability with respect to the size of the reference database used for completion: the output of Meneco remains reasonable in terms of size ( for an a posteriori manual curation ) , with a reasonable loss of impact on the restoration of biomass production compared to parsimonious approaches . To further evaluate the quality of the output of Meneco , we completed the 3 , 600 degraded GEMs of our benchmark derived from the iJ904 E . coli metabolic network by 7 , 200 ( 2*3 , 600 ) additional degraded GEMs of the iAF1260 [32] and iJO1366 [33] E . coli networks . The Meneco tool was applied to complete each of the 10 , 800 degraded GEMs by using the networks prior to degradation as a reference database . The main motivation for changing the reference database was to be able to analyze how the classification of reactions into essential , alternative and blocked with respect to biomass production evolved with the completion process . As a second case study , we focused on the reconstruction of a metabolic network for E . mutabilis , a photosynthetic protist and primary producer in acidic aquatic environments . Despite important roles of E . mutabilis in those ecosystems and despite the fact that it has often been considered as an indicator species for acid mine drainages , this organism has been poorly described so far . Due to their complexity and size , no genome sequences are available so far within the genus Euglena . Therefore , we used assembled transcript sequences as described in [18] and annotated them by similarity search against enzyme reference sequences from MetaCyc . We used a stringent similarity threshold for function assignation to reduce as much as possible the rate of false positive reactions included in the initial network that was subsequently completed by Meneco . This ensured that the reconstruction of the E . mutabilis metabolic network would be highly conservative , avoiding the inclusion of reactions based on false annotations . The major drawback of this approach was the resulting high rate of missing reactions in the initial draft network , which did not allow Meneco to restore the production of all targets at once . The 72 targets are listed in Supplementary files S6 . Of those targets , 54 were defined according to [19] who showed that they were secreted or accumulated in E . mutabilis cultured in a minimal mineral medium . The 17 other targets were myristic acid , shown to be present in the E . mutabilis membrane [36] , and the remaining proteinogenic amino acids , nucleotides , and chlorophyll A and B . Running Meneco with the 8 mineral seeds and |Light| ( Supplementary files S6 ) could propose a completion solution only for UREA and OXYGEN-MOLECULE among the 72 targets . These two metabolites could be produced only because they were the products of a single reaction involving only mineral reactants , a situation which was not biologically satisfying .
The main characteristic of Meneco is that it relies on a topological description of the notion of producibility . The qualitative approximation of the topological notion of producibility is more robust than stoichiometry-based notions , which are often forced to find more complete solutions that additionally fulfill the stoichiometric constraints . The logical formalization of the topological producibility allows users of Meneco to benefit from the performance of recent solvers for combinatorial problems based on ASP technologies . Meneco relies on these efficient solvers for the completion of degraded draft metabolic networks built , for instance , from NGS data . The other characteristic of the Meneco tool is that it is very flexible with respect to the definition of seeds ( composition of the medium and potentially some cofactors ) and targets ( compounds whose producibility has to be restored ) . This flexibility appears to be crucial to investigate draft GEMs produced from NGS technologies , as illustrated in two applications . Most gap-filling methods solve optimization problems over a search space whose size grows exponentially with the size of the reference database from which the reactions are taken . Two different strategies can then be used to explore the search space and compute solution sets to a gap-filling problem: either a parsimonious bottom-up strategy which enriches the draft metabolic network until the targeted properties are satisfied [8 , 10 , 27] , or a top-down approach starting from all available information and removing reactions without added-value to the solution of the problem [15 , 23 , 24] . Bottom-up methods often report few solutions and may miss alternative ones while top-down approaches capture more solutions but are computationally demanding , and sometimes require sampling of the solution space . Therefore , families of gap-filling methods can be classified with respect to four characteristics: ( i ) the set of compounds whose producibility should be restored; ( ii ) the definition of producibility they rely on; ( iii ) the criteria they optimize; ( iv ) the number of solution sets they return . Meneco checks the same topological criteria for producibility as the subset-minimality top-down method [15] , but it is able to enumerate all solution sets by using the parsimony criterion of the GapFill family . This allows an exhaustive computation of the family of solutions . From a biological point of view , having a method that enables the completion of a metabolic network without the need to know the real concentration of metabolites is a real advantage . It enables a completion of the metabolic graph not only for quantified or estimated compounds , but also for those identified by qualitative measurements . Moreover , having access to an exhaustive enumeration of the possible solutions allows researchers to choose the best one among them , instead of having only a subset of solutions without knowing if this subset is representative of the entire solution space or not . Alternative qualitative semantics could have been used to assess the producibility of a metabolic compound . In [26 , 37] , Cottret et al . introduced refined semantics for producibility taking into account the impact of cycles on the production of metabolites . This alternative definition can be viewed as an over-approximation of FBA-consistent networks , and appears to be very useful for the efficient computation of precursor sets for metabolic targets . However , experiments consisting in encoding this alternative definition in the Meneco framework and performing gap-filling of the same 10 , 800 degraded networks evidenced that this alternative definition of producibility is not constrained enough to capture essential reactions . Indeed , running our benchmark on the iJR904 network with that definition of producibility returns a set of solutions containing on average only 59 . 5% of the essential reactions ( S1 Files ) . This can be explained by self producing cycles which can occur with that definition . Our experiments using the E . coli benchmark evidence that , in half of the cases , Meneco ( and GapFill ) fail to recover enough alternative reactions to restore the biomass producibility of the network . This bottleneck can be explained by the parsimony criterion used by both tools since they identify sets with a minimal number of reactions allowing either to simultaneously restore the topological producibility of all targets ( Meneco ) or to enable the production of individual targets ( GapFill ) . However , the identification of alternative routes to produce a targeted set of compounds is crucial to have a global understanding of species metabolic capability , as soon as essential reactions have been properly identified and validated . In order to improve the Meneco tool , we plan to study the impact of additional topological criteria that may take into account larger pathways . The difficulty will be to select relevant metrics in order to extend the search space while sufficiently constraining the search to be able to track the complete solution space . Indeed , with most of the methods based on extended scores and criteria for stoichiometry-based formalisms [10 , 15] , the space or compatible sets of reactions may become intractable , especially when many targets are considered together . In this case , most methods rely on a sampling of the solution space , which may introduce biases and errors . Alternatively , other metrics can be developed to improve the choice of reactions by the gap-filling methods . Scores based on likelihood value computations have been introduced in [10] to improve GapFill approaches with genomic information about alternative functions for genes . In the future , it will be interesting to adapt such scores to the Meneco tool and measure their impact on the functional classification of reactions in the reconstructed network . As stated by Satish Kumar et al . in [8] , “clearly , the role of a gap-filling method is to simply pinpoint a number of hypotheses which need to subsequently be tested” . In this line of research , the Meneco tool constitutes a flexible framework to pinpoint hypotheses in the context of large-scale datasets applied to newly investigated organisms . Meneco is a versatile tool to complete draft GEMs and to suggest relevant reactions with respect to the response of the system to environmental perturbations . Importantly , it does not aim at providing a complete functional network , but rather at pointing out essential reactions and some of the alternative ones , which are crucial to explain the system response . It can be then combined with refined stoichiometry-based analyses and gap-filling methods to produce functional networks . In this sense , we promote Meneco as a tool to be used as an intermediary step within a workflow consisting of ( i ) producing a draft GEM [5 , 38] , ( ii ) parsimonious gap-filling based on metabolite profiles or RNA-seq datasets with Meneco , ( iii ) refinement of the model with stoichiometric-based approaches relying on additional data ( [14] [13] ) , and ( iv ) manual curation process of metabolic networks described in [1] . From a biological point of view , the examples show that Meneco cannot replace manual curation and network analysis , but it may provide a flexible tool to aid this process . Analyses are fast , easy to implement , and invaluable because they enable biologists to focus their attention on a few highly interesting compounds and reactions without making a priori assumptions . | In the era of fast and massive genome sequencing , one challenge is to transform sequence information into biological knowledge . Reconstructing metabolic networks that include all biochemical reactions of a cell is a way to infer reactions from genomic data . Unfortunately , those data are usually incomplete , poorly annotated , and missing reactions create gaps in the metabolic networks . Here we introduce Meneco , a tool dedicated to the parsimonious gap-filling of metabolic networks . Unlike other tools , Meneco allows using sparse data ( missing stoichiometries ) and draft metabolic networks to suggest reactions to fill gaps in the networks . Subsequently , we apply it to two biological case studies and show that the flexibility of Meneco enables it to be adapted to a variety of research questions and types of available data . We show that Meneco performs better than reference tools with respect to large-scale heterogeneous reference database and with respect to the recovery of important reactions in highly degraded networks . Specifically , it allowed the analysis of two interacting metabolic networks and the reconstruction of the first metabolic network of Euglena mutabilis . | [
"Abstract",
"Introduction",
"Results",
"Discussion"
] | [
"metabolic",
"networks",
"enzymology",
"database",
"searching",
"reactants",
"genomic",
"databases",
"plant",
"science",
"metabolites",
"network",
"analysis",
"genome",
"analysis",
"enzyme",
"metabolism",
"photosynthesis",
"plants",
"enzyme",
"chemistry",
"research",
"and",
"analysis",
"methods",
"computer",
"and",
"information",
"sciences",
"biological",
"databases",
"chemistry",
"algae",
"biochemistry",
"plant",
"biochemistry",
"database",
"and",
"informatics",
"methods",
"genetics",
"biology",
"and",
"life",
"sciences",
"chemical",
"reactions",
"physical",
"sciences",
"genomics",
"metabolism",
"computational",
"biology",
"organisms"
] | 2017 | Meneco, a Topology-Based Gap-Filling Tool Applicable to Degraded Genome-Wide Metabolic Networks |
Sri Lanka was one of the first countries to initiate a lymphatic filariasis ( LF ) elimination program based on WHO guidelines . The Anti-Filariasis Campaign provided 5 annual rounds of mass drug administration ( MDA ) with diethylcarbamazine plus albendazole in all 8 endemic districts from 2002–2006 . Microfilaremia ( Mf ) prevalences have been consistently <1% in all sentinel and spot-check sites since 2006 , and all evaluation units passed school-based transmission assessment surveys ( TAS ) in 2013 . We previously reported results from comprehensive surveillance studies conducted in 2011–2013 that documented low-level persistence of Wuchereria bancrofti in 19 high risk areas in 8 endemic districts . We now present results from repeat surveys conducted 3 to 4 years later in 6 areas that had the strongest LF signals in the prior study . The surveys assessed prevalence of filarial antigenemia ( CFA ) and Mf in communities , CFA and anti-filarial antibody in school children ( ages 6–8 ) , and filarial DNA in Culex mosquitoes ( molecular xenomonitoring , MX ) . Three study areas had significantly improved infection parameters compared to the prior study , but three other areas had little change . MX was more sensitive for detecting W . bancrofti persistence , and it was a better predictor than other parameters . Adult males accounted for more than 80% of infections detected in the study . These results suggest that W . bancrofti transmission was near the break point in some of the areas studied in 2011–13 . LF is likely to decline to zero without further intervention in these areas , while other areas may require further intervention . Long term surveillance may be needed to verify W . bancrofti elimination in areas like Sri Lanka with efficient transmission by Culex . Test and treat or other programs targeting adult males plus bed net promotion may be more effective than MDA for clearing remaining hotspots of transmission in Sri Lanka .
Lymphatic filariasis ( LF , caused by the filarial nematodes Wuchereria bancrofti , Brugia malayi , and B . timori ) , is a major public-health problem in many tropical and subtropical countries . The global program to eliminate Lymphatic Filariasis ( GPELF ) has made significant progress by providing more than 6 billion treatments with antifilarial medications to more than 800 million people in some 60 countries between 2000 and 2015 [1] . This mass drug administration ( MDA ) program has cured millions of infections and prevented millions of new clinical filariasis cases [1–3] . Countries with successful MDA programs are now looking for additional guidance on stopping MDA and on post-MDA surveillance beyond WHO current guidelines[1 , 4 , 5] that rely heavily on testing school aged children for filarial antigenemia as a means of demonstrating that transmission of new infections has been interrupted . While such “transmission assessment surveys” ( TAS ) can be a useful surveillance tool [4 , 6] , they have not been adequately validated as an indicator for interruption of LF transmission at the evaluation unit or country level . Indeed , prior studies by our group have shown that TAS was not sensitive for detecting ongoing transmission of W . bancrofti in Sri Lanka [7] , and this is likely to be true in many other settings . Lymphatic filariasis has been endemic in Sri Lanka for hundreds of years [8–11] . The country’s Anti Filariasis Campaign ( AFC , established in 1947 ) implemented control activities over many years that succeeded in reducing infection prevalence to low levels by 1999 . After providing mass drug administration of diethylcarbamazine ( DEC ) for three years starting in 1999 , the AFC provided five annual rounds of MDA with diethylcarbamazine ( DEC ) plus albendazole in all 8 endemic districts ( implementation units , IU ) between 2002 and 2006 [2 , 12–14] . The AFC conducted post-MDA surveillance activities according to WHO guidelines , and all evaluation units in endemic districts easily passed TAS in 2013 [7] . Based on this and other considerations , WHO recognized that Sri Lanka had eliminated LF as a public health problem in 2016 , but recommended that the country continue treatment interventions with high population coverage and post-MDA surveillance in isolated foci with evidence of ongoing transmission [1 , 15 , 16] . We assessed the status of W . bancrofti in Sri Lanka with comprehensive post-MDA surveillance in 19 Public Health Inspector areas that were considered to be at risk for persistent infection . Comprehensive surveillance comprised community surveys for circulating filarial antigenemia ( CFA ) and microfilaremia ( Mf ) , school surveys for CFA and antifilarial antibodies , and systematic sampling of Culex quinquefasciatus for the presence of filarial DNA ( molecular xenomonitoring or MX ) [7] . All 19 sentinel areas studied had evidence for persistent W . bancrofti , but some areas had stronger signals than others . Based on results of that study , we suggested revised endpoint targets for filariasis elimination programs in areas with Culex transmission based on upper 95% confidence limits as follows: CFA <2% , antibody prevalence in primary school children <5% , and filarial DNA prevalence in gravid , semigravid , or fed Culex mosquitoes <1% [7] . In the present study we have repeated comprehensive surveillance in 6 areas with the strongest signals for W . bancrofti persistence in 2011–2013 to determine whether there was evidence for improvement or worsening of infection parameters 3 to 4 years after the prior study .
The study protocol was reviewed and approved by institutional review boards at Washington University School of Medicine , University of Kelaniya and at the Ministry of Health in Sri Lanka . Printed copies of participant information sheets ( PIS ) and written consent forms were provided to participants ( or to parents/guardians ) in Sinhalese , Tamil and English . Written consent was obtained from adults; participation of minors required written consent from at least one parent or guardian plus assent by the child/minor . The study was performed in four Public Health Inspector ( PHI ) areas , one Public Health Field Officer ( PHFO ) area and one in Colombo municipality area that had evidence of persistent LF in a post-MDA surveillance study of 19 areas that was conducted in 2011–13 [7] ( Fig 1 ) . PHIs are sub-district health administrative units with populations in the range of 10 , 000–30 , 000 that are comprised of smaller units called Public Health Midwife ( PHM ) areas . The current study was performed three to four years after the last evaluation . No treatment for filariasis was provided in this interval in 4 of these PHIs . One round of MDA with DEC plus albendazole was provided in two of the PHIs ( Unawatuna and Ambalangoda in Galle district ) in 2014 . Field procedures were the same as those previously described [7] . Briefly , field teams for collection of demographic information and blood samples consisted of a medical officer , a Public Health Inspector , a data entry operator , a phlebotomist , and one or two assistants . 1 . 5 mm x 2 . 0 mm , blue and 21 G x 1 . 8 mm pink single use contact-activated BD-microtainer lancets ( Fisher scientific , Pittsburgh , PA ) were used for blood collection in community and school surveys , respectively . Blood samples were collected during the day . Approximately 300 to 400 μl of blood was collected by finger prick from each study participant into an EDTA coated blood collection vial ( Fisher Scientific ) . Preprinted barcode labels were used to link samples to participant records . Samples were transported to the central Antifilariasis Campaign ( AFC ) laboratory in Colombo in coolers . Plasma was separated by centrifugation from blood samples from school surveys and stored at -80 C for antibody testing . Survey methods were the same as previously described [7] . Briefly , area maps , census information ( the number of houses and the number of households , number of schools , and the number of primary grade children ) were obtained from census records , voter lists , and from school principals and administrators [17] . Community surveys sampled approximately 500 participants ( ages 10–70 years ) in approximately 125 households per PHI/PHFO area ( range 127–172 HH ) . Most houses in the study area have 3 or 4 residents in this age range . Young children tested in school surveys were too young for inclusion in the community surveys ( no overlap ) . Systematic sampling was used for household and mosquito sampling in each PHM within the PHI area . The number of houses/households needed for each community survey ( 125 ) was divided by number of PHMs in the PHI to get the number of houses to be sampled in each PHM . That number was divided by 4 to get the number of houses to be sampled in each quadrant in each PHM area . The sampling interval for houses was calculated by dividing the number of houses that were to be sampled in that PHM quadrant . Households from all quadrants in the PHM were enrolled . To maintain consistency in sampling and to obtain geographically dispersed samples , only 4 subjects ≥ 10 years were enrolled per household with equal preference for males and females . School surveys were performed in all schools that served the sentinel area . Finger prick blood was collected from primary grade school children ( grades 1 and 2 , age 6–8 ) and community participants for antigen and antibody testing . Circulating filarial antigenemia ( CFA ) was detected in finger prick blood samples with a rapid format card test ( BinaxNOW Filariasis , Alere Inc . , Scarborough , ME ) according to the manufacturer’s instructions . Cards were read visually at 10 minutes . Antigen testing was performed within 24 hr of blood collection . IgG4 antibodies to recombinant filarial antigen Bm-14 in human plasma were detected by microplate ELISA ( Filariasis CELISA , Cellabs Pty Ltd , Brookvale , NSW , Australia ) as previously described [18] . Plasma samples were tested in a single well per sample and all positive and borderline tests with OD values >0 . 35 were retested on a different day to confirm their positivity . Samples with OD values consistently >0 . 35 were considered to be positive for antibodies to Bm14 . Persons with a positive filiarial antigen test had night blood testing to detect microfilaremia ( Mf ) as previously described . Briefly , finger prick blood collected between 9 pm and 12 midnight was used to prepare three-line blood smears ( 60 μl total volume of blood tested ) that were dried , fixed , stained with Giemsa , and examined by microscopy for the presence of Mf . Each stained slide was read by a single experienced microscopist who recorded the absence or presence of Mf and Mf count . Culex quinquefasciatus were collected with CDC gravid traps ( Model 1712 , John W . Hock Company , Gainesville , FL ) as previously described [7 , 19] . Briefly , traps were placed outside houses in shaded areas in all quadrants of each PHM to ensure proportional sampling from all areas in each PHI . Trapped mosquitoes were sorted , dried at 950 C for 1 hr . , and placed in tubes for later molecular testing . Four pools of twenty fed , gravid , or semigravid female Cx . quinquefasciatus were tested from each of 50 trapping locations per PHI . Extraction of DNA from mosquitoes and detection of W . bancrofti DNA by qPCR were performed at the AFC central laboratory as previously described [20] . Demographic information was collected and entered into BLU phones ( BLU products , Miami , FL ) using preloaded survey forms with LINKS data collection software https://www . linkssystem . org . Cell phones are equipped with global positioning system ( GPS ) capability , and GPS coordinates were captured at each surveyed house and mosquito trap location . Enrollment forms collected information on age , gender , consumption of antifilarial medications during the 2000–2006 MDA , bed net use last night , and clinical signs of lymphedema ( all self-reported ) . Participant data and specimens were linked to laboratory test results with preprinted barcode labels . Deidentified , cleaned data were transferred into Microsoft Excel ( Microsoft Corp . , Redmond , WA ) for analysis . Households included in population surveys and mosquito trapping sites were mapped using ArcGIS 10 . 2 . 1 ( ESRI , Redlands , CA ) . Chi-squared or Fisher’s exact tests were used to assess the significance of differences in filariasis parameters ( prevalence of surveyed persons positive for antigenemia and antibody and percentages of mosquito pools that contained filarial DNA ) . Prevalence of filarial DNA in mosquitoes ( maximum likelihood and 95% CI ) were estimated using Poolscreen 2 . 02 software [21 , 22] . Filarial DNA prevalence values in mosquitoes were considered to be significantly different if there was no overlap in the 95% CI values for the two samples . Correlations between human and mosquito infection parameters were assessed with the Spearman rank test . Graphs were produced with GraphPad Prism 7 software ( La Jolla , CA ) .
Six areas in 5 districts were resurveyed for W . bancrofti infection parameters between January 2015 and February 2017 . LF surveillance periods for sample collections were in Peliyagodawatta ( Oct . , Nov . , 2011 and Jan . , Feb . , 2015 ) ; Kalutara North ( Sept . , Oct . , 2011 and Oct . , 2015 ) ; Ambalangoda ( Nov . , Dec . , 2011 and March , May 2015 ) ; Unawatuna ( Nov . , Dec . , 2011 and July , Aug . , 2015 ) ; Weligama ( June , Sept . , 2012 and Nov . , Dec . , 2015 ) ; Borella ( April 2013 and Oct . , 2016 to Feb . , 2017 ) . A total of 5350 people from the six sentinel sites participated in the study . This total included 3123 people in community surveys ( ages 10–70 , mean age 38 years , 42% males ) and 2227 children ( age 6–8 ) in school surveys . Details for enrollment in community surveys by sentinel site are provided in Table 1 . Few filarial lymphedema cases ( 34 of 3123 , 1 . 1% ) were identified during the survey . Reported bed net use was moderate to high ( range 49% to 70% ) in all PHIs studied except Borella . Survey results are summarized in Table 2 . CFA prevalences were lower than 2% in all areas , but upper confidence limits for CFA were greater than 2% in the two PHI areas in Galle district . All CFA tests were negative in two PHIs . CFA prevalence was higher in males than in females in PHIs with at least one positive card test [13 of 869 ( 1 . 5% , 0 . 8–2 . 5 CI ) vs . 3 of 1211 ( 0 . 2% , 0 . 1–0 . 7 CI ) , P = 0 . 003] . CFA prevalence was higher in adults ( age ≥ 18 ) than in children ( ages 10–17 ) in the community surveys [15 of 1731 ( 0 . 9% , 0 . 5–1 . 4 CI ) vs . 1 of 349 ( 0 . 3% , 0 . 05–1 . 60 CI ) , P = 0 . 2] . Ten of 16 persons with positive CFA tests were over the age of 50 ( range 54–69 ) ; eight of 10 ( 80% ) CFA positives in this age group were males . Mf prevalences were well under 1% in all surveyed PHI areas . Three of 16 persons with positive CFA tests in the community surveys were also Mf positive ( range 2–9 Mf count in 60 μl ) with one each from Ambalandgoda , Unawatuna and Weligama . CFA prevalence was low in all PHIs ( Table 2 ) , but the upper 95% CI exceeded 2% only in the Unawatuna PHI ( Galle district ) . Only one of 5 CFA-positive children was Mf-positive ( 1 mf/60 μl ) . Antibody prevalences were higher in the 3 PHI areas in the Southern province ( upper CI close to or higher than 5% ) than in the other sentinel sites ( Table 2 ) . No child had a positive antibody test in Borella . MX results are summarized in Table 3 . Filarial DNA was detected in mosquitoes in all of six PHI areas . However , filarial DNA prevalence exceeded the target ( upper CI > 1% ) in three PHIs ( Unawatuna and Ambalangoda in Galle district and Weligama in Matara district ) . Many trap locations were positive for mosquitoes with filarial DNA in Unawatuna , Ambalangoda and Weligama ( Table 4 ) . The percentages of positive mosquito trap sites in PHIs were not significantly correlated with percentages of houses with at least one CFA positive person ( Table 4; Spearman rank correlation , r = 0 . 5 , P = 0 . 2 ) . Analysis on few data points in 6 areas may have caused this poor correlation . The maps in Fig 2A–2D show locations for households surveyed for CFA and mosquito trapping sites in 4 PHIs . Many more positive mosquito trapping locations were identified than positive households . Positive mosquito trap sites were widely dispersed in three PHIs where approximately 50% of trap locations yielded mosquitoes with filarial DNA . In contrast , positive mosquito trap sites in the Kalutara North were concentrated in the southern part of that PHI . Community antibody testing was performed in Unawatuna and Weligama PHI areas that had evidence of persistent W . bancrofti infections and transmission in the baseline surveys . Community ( age ≥10 ) antibody prevalence was very high in Unawatuna ( 168/506 , 33% , 95% CI 29–37% ) and in Weligama ( 166/501 , 33 . 1% , 95% CI 29–37% ) , and these values were much higher in adults ( age ≥18 ) than those in children ( Fig 3 ) . Only a small number of people with positive antibody tests were positive for CFA ( 6/168 in Unawatuna and 1/166 in Weligama ) . Overall anti-filarial antibody prevalence was significantly higher in males than females ( 42% vs 26% , P = 0 . 0001 , combined results from Unwatuna and Weligama ) ( Fig 3 ) . Results from surveys conducted in Peliayagodawatta in 2008 , 2011 and 2015 are summarized in Table 5 . W . bancrofti infection parameters spontaneously improved over time in this area , and most of these changes were statistically significant . In addition , the trend was consistent over time with greater reductions from baseline 2008 values in 2015 than in 2011 . Longitudinal data on LF infection parameters from all six PHIs are shown in Fig 4 . CFA and MX results improved significantly between 2011 and 2015 in Peliyagodawatta and Kalutara North and between 2013 and 2017 in Borella . There was also a slight downward trend in some LF parameters in Unawatuna , Ambalangoda and Weligama , but these changes were not statistically significant even though the government provided one round of MDA with DEC plus albendazole in Unawatuna and Ambalangoda during this interval . Filarial DNA prevalences in mosquitoes were high ( MLE > 0 . 25% , upper CI > 1% ) in 2015 in these PHIs . CFA prevalence was significantly lower in Unawatuna and Ambalangoda community participants ( n = 1043 , age ≥10 ) who reported bed net use ( 0 . 60% vs 2 . 42% in non-users , P = 0 . 01 ) . There was no association between anti-filarial antibody and bed net use when all ages were considered in Unawatuna and Weligama , but antibody prevalence in children aged 10–17 were significantly lower in bed net users than in nonusers ( 8/96 , 8 . 3% vs . 12/62 , 19 . 4% , P = 0 . 04 ) .
This study has provided interesting new data on changes in LF parameters post-MDA . LF elimination requires reduction of infection parameters to levels that cannot sustainably support transmission . This does not mean that all measures of LF must be zero , and indeed we found evidence of low-level persistence of LF in all 6 PHI areas that were restudied in 2015–16 . Breakpoints for LF transmission ( where filariasis parameters have been reduced below levels required for sustained transmission ) are poorly defined , and they depend on many factors that are difficult to measure and may vary widely between and within endemic regions . We have proposed targets for LF elimination programs [7 , 23] , and the current study attempted to “ground truth” these targets . While the time intervals and number of study sites were not sufficient to rigorously prove the validity of the targets , our results suggest that they are in the right range and also feasible for measurement by national programs . Results from Sri Lanka are likely to apply to many other LF-endemic areas with transmission by Culex mosquitoes . Long-term post-MDA surveillance will be needed to verify LF elimination in areas like Sri Lanka that have highly competent vectors . As in our prior study , data from 2015–16 again show that MX and antibody testing of school children are more sensitive than antigen testing of school children for detecting low-level persistence of LF in post-MDA settings . Results for 3 parameters measured in this study and in 2011–13 ( community CFA , antibody prevalence in school children , and prevalence of filarial DNA in mosquitoes ) support the provisional target values for LF elimination programs ( upper 95% CI values of 2% for community CFA , 5% for antibody in school children , and 1% for MX ) , and areas that failed to meet one of the targets often failed to meet the others . Results from Peliyagodawatta suggest that the community CFA target of 2% in the post-MDA setting may be too conservative , because the prevalence in that study site declined from 3 . 8 in 2008 to 0 . 4% in 2015 without intervention . One weakness of school-based TAS as currently performed is that signals from focal high infection areas are often diluted when evaluation units are large . Evaluation Units with populations of one million or more are commonly employed by LF elimination programs in Asia [1 , 4] . The ideal EU size is not known , but reducing the population for EUs to 200 , 000 or lower should be more sensitive for detecting persistence or resurgence of LF than the currently recommended ceiling of 2 million . Large EUs were needed to reduce surveillance costs when the cost of CFA tests was high . However , recent changes have reduced these costs , and this may make it feasible to reduce EU size and perform more TAS . A recent study has modeled effects of EU size and population on sensitivity for detecting ongoing hotspots of transmission [24] . It is not clear whether this information can be translated into changes in policy or practice that are feasible for use by national LF elimination programs . While school-based TAS with a point of care antigen test is a convenient way to sample a sentinel population for recent infections , this approach is less sensitive than the other parameters that we tested . The strategy of sampling sentinel populations does not work well if the sentinels are at low risk for infection . Antigen data in this study and in our prior study show that adult males have much higher filarial infection prevalence than other groups in Sri Lanka , and they represented the bulk ( >80% ) of the residual reservoir of infection in PHI areas surveyed in this study . Is this because they have more exposure to infective mosquitoes , higher susceptibility to infection , and lower compliance with MDA , or a combination of these factors ? A TAS that focuses on high-risk adult males to assess the persistent reservoir of infection might be a more effective tool for post-MDA surveillance than school-based TAS that aims to detect recent infections . Antifilarial antibody test results ( reflecting both recent and past filarial infections ) also showed age-related increases in prevalence , and antibody prevalence was much higher in males than females . These results underscore gender and age differences in LF infection and exposure in Sri Lanka . A post-MDA surveillance study in American Samoa found similar results with increased infection prevalence in adult males [25] . Our finding of high antibody prevalence in adults in areas that were close to LF elimination suggest that testing adults with this antibody test ( IgG4 antibodies to recombinant filarial antigen Bm14 ) has little value as a post-MDA surveillance tool . Forty-one of 1 , 700 ( 2 . 3% ) children tested in 5 PHI areas with one or more child with a positive antibody test had positive antibody tests , and 6 of these children had positive antigen tests . These children were born after Sri Lanka’s national MDA program was completed . Prior studies have shown that persons with positive antifilarial antibody tests have an increased risk for developing microfilaremia during follow-up [26 , 27] , so we recommend presumptive treatment for children with positive antibody tests . AFC currently provides antifilarial treatment to persons with microfilaremia or positive antigen tests according to WHO guidelines . The longitudinal results in this study are especially interesting . They suggest that filariasis parameters in Peliyagodawatta in 2008 and in Kalutara North and Borella in 2011–13 were already below transmission breakpoints , and LF appears to be on a glide path to elimination in these areas . On the other hand , results from Unawatuna , Ambalangoda , and Weligama suggest that transmission is ongoing in these areas and that they will require further intervention . Of the various parameters measured , the filarial DNA prevelance in mosquitoes ( as assessed by MX ) seems to have been the best predictor for LF persistence . Positive MX results in areas where little or no infection was detected in humans ( Borella , Kalutara North , and Peliyagodawatta ) is intriguing . It is likely that there are infected persons in these communities who were non-compliant with MDA in the past and also not sampled in our community surveys . Mosquitoes do not ask permission when they conduct night blood sampling , and this probably accounts for the enhanced sensitivity of MX for detecting persistent infections relative to other modalities . Results from this study provide useful insights regarding approaches for clearing up LF transmission hotspots in post-MDA settings like those in southern Sri Lanka . Resumption of MDA is not an efficient option when human infection prevalence is very low; MDA will not benefit the vast majority of people who are uninfected , and the program is likely to miss most persons with persistent infection who have been non-compliant with MDA in the past . The Sri Lanka AFC provided one round of MDA with DEC plus albendazole in 2014 ( prior to this study ) , and also provided MDA to selected areas within Galle district in 2015 and 2016 . We believe that instead of focusing on the percentage of the population that can be reached with MDA ( population coverage ) , programs should focus on how to optimize treatment of infected persons ( worm coverage ) . Since more than 80% of those with antigenemia in the present study and approximately 65% of those with antigenemia in the study published in 2014 were adult males , a “test and treat” program or other approaches that focus on adult males might result in higher worm treatment coverage than population-based MDA . Population antigen data and antibody data from children in Unawatuna and Weligama PHIs in Galle and Matara districts point to a potential protective effect of bed nets for Culex-transmitted LF in Sri Lanka . This was an unexpected finding , because bed nets are considered to be more important for LF control in settings with anopheline transmission [28] . Bed nets are popular in Sri Lanka , because they help to reduce the mosquito nuisance and because they may provide some protection against dengue virus infection . Additional promotion of bed nets or a focused government subsidy program for bed nets in selected areas with persistent LF may help to clear remaining LF hotspots . | Lymphatic Filariasis ( LF , also known as “elephantiasis” ) is a disabling and deforming tropical disease caused by parasitic worms that are transmitted by mosquitoes . The Sri Lankan Anti-Filariasis Campaign provided 5 annual rounds of mass drug administration ( MDA ) with diethylcarbamazine and albendazole between 2002 and 2006 in all endemic areas , and this reduced infection prevalence to very low levels . Post-MDA surveillance conducted by our group in 19 sentinel sites in 8 endemic districts in 2011–2013 revealed evidence of persistent LF infection in all study sites . The present paper reports results of repeat assessments conducted 3–4 years later in 6 areas with high signals in the prior study . LF parameters were significantly improved in 3 areas where LF appears to be on a glide path to elimination . However , LF infection parameters remained high in 3 areas , and further work will probably be required to interrupt transmission in these areas . Molecular xenomonitoring ( to detect filarial DNA in mosquito vectors ) was especially sensitive for detecting persistent LF in Sri Lanka , and this may also be true in other areas with Culex transmission . Our results suggest that test and treat or other programs targeting adult males plus expanded bed net use may be helpful for clearing up remaining LF hotspots . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [
"invertebrates",
"medicine",
"and",
"health",
"sciences",
"immune",
"physiology",
"body",
"fluids",
"education",
"sociology",
"immunology",
"geographical",
"locations",
"social",
"sciences",
"parasitic",
"diseases",
"animals",
"filariasis",
"antibodies",
"insect",
"vectors",
"public",
"and",
"occupational",
"health",
"immune",
"system",
"proteins",
"infectious",
"diseases",
"wuchereria",
"bancrofti",
"proteins",
"wuchereria",
"sri",
"lanka",
"schools",
"disease",
"vectors",
"insects",
"arthropoda",
"people",
"and",
"places",
"biochemistry",
"helminth",
"infections",
"mosquitoes",
"eukaryota",
"blood",
"asia",
"anatomy",
"physiology",
"nematoda",
"biology",
"and",
"life",
"sciences",
"species",
"interactions",
"organisms"
] | 2017 | Reassessment of areas with persistent Lymphatic Filariasis nine years after cessation of mass drug administration in Sri Lanka |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.